Datasets:

billxbf

/

arxiv_dump

like 1

2103.02728

Cosmin Badea

Cosmin Badea, Gregory Artus

Morality, Machines and the Interpretation Problem: A Value-based, Wittgensteinian Approach to Building Moral Agents

In: Bramer, M., Stahl, F. (eds) Artificial Intelligence XXXIX. SGAI-AI 2022. Lecture Notes in Computer Science(), vol 13652. Springer, Cham (2022)

10.1007/978-3-031-21441-7_9

cs.AI cs.CL cs.CY

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We present what we call the Interpretation Problem, whereby any rule in symbolic form is open to infinite interpretation in ways that we might disapprove of and argue that any attempt to build morality into machines is subject to it. We show how the Interpretation Problem in Artificial Intelligence is an illustration of Wittgenstein's general claim that no rule can contain the criteria for its own application, and that the risks created by this problem escalate in proportion to the degree to which to machine is causally connected to the world, in what we call the Law of Interpretative Exposure. Using game theory, we attempt to define the structure of normative spaces and argue that any rule-following within a normative space is guided by values that are external to that space and which cannot themselves be represented as rules. In light of this, we categorise the types of mistakes an artificial moral agent could make into Mistakes of Intention and Instrumental Mistakes, and we propose ways of building morality into machines by getting them to interpret the rules we give in accordance with these external values, through explicit moral reasoning, the Show, not Tell paradigm, the adjustment of causal power and structure of the agent, and relational values, with the ultimate aim that the machine develop a virtuous character and that the impact of the Interpretation Problem is minimised.

2023-02-08

2103.02800

Zejian Liu

Zejian Liu, Gang Li and Jian Cheng

Hardware Acceleration of Fully Quantized BERT for Efficient Natural Language Processing

Design, Automation & Test in Europe (DATE) 2021

cs.AR cs.CL

http://creativecommons.org/licenses/by/4.0/

BERT is the most recent Transformer-based model that achieves state-of-the-art performance in various NLP tasks. In this paper, we investigate the hardware acceleration of BERT on FPGA for edge computing. To tackle the issue of huge computational complexity and memory footprint, we propose to fully quantize the BERT (FQ-BERT), including weights, activations, softmax, layer normalization, and all the intermediate results. Experiments demonstrate that the FQ-BERT can achieve 7.94x compression for weights with negligible performance loss. We then propose an accelerator tailored for the FQ-BERT and evaluate on Xilinx ZCU102 and ZCU111 FPGA. It can achieve a performance-per-watt of 3.18 fps/W, which is 28.91x and 12.72x over Intel(R) Core(TM) i7-8700 CPU and NVIDIA K80 GPU, respectively.

2021-03-05

2103.02845

Xingyu Chen

Xingyu Chen, Yufeng Liu, Chongyang Ma, Jianlong Chang, Huayan Wang, Tian Chen, Xiaoyan Guo, Pengfei Wan, Wen Zheng

Camera-Space Hand Mesh Recovery via Semantic Aggregation and Adaptive 2D-1D Registration

CVPR2021

CVPR2021

cs.CV

http://creativecommons.org/licenses/by-nc-nd/4.0/

Recent years have witnessed significant progress in 3D hand mesh recovery. Nevertheless, because of the intrinsic 2D-to-3D ambiguity, recovering camera-space 3D information from a single RGB image remains challenging. To tackle this problem, we divide camera-space mesh recovery into two sub-tasks, i.e., root-relative mesh recovery and root recovery. First, joint landmarks and silhouette are extracted from a single input image to provide 2D cues for the 3D tasks. In the root-relative mesh recovery task, we exploit semantic relations among joints to generate a 3D mesh from the extracted 2D cues. Such generated 3D mesh coordinates are expressed relative to a root position, i.e., wrist of the hand. In the root recovery task, the root position is registered to the camera space by aligning the generated 3D mesh back to 2D cues, thereby completing cameraspace 3D mesh recovery. Our pipeline is novel in that (1) it explicitly makes use of known semantic relations among joints and (2) it exploits 1D projections of the silhouette and mesh to achieve robust registration. Extensive experiments on popular datasets such as FreiHAND, RHD, and Human3.6M demonstrate that our approach achieves stateof-the-art performance on both root-relative mesh recovery and root recovery. Our code is publicly available at https://github.com/SeanChenxy/HandMesh.

2022-04-01

2103.02854

Dexter Neo

Vassilios Vonikakis, Dexter Neo, Stefan Winkler

Morphset:Augmenting categorical emotion datasets with dimensional affect labels using face morphing

in Proc IEEE International Conference on Image Processing (ICIP), Anchorage, Sep.2021

2021 IEEE International Conference on Image Processing (ICIP), 2021

10.1109/ICIP42928.2021.9506566

cs.CV cs.AI

http://creativecommons.org/licenses/by-nc-nd/4.0/

Emotion recognition and understanding is a vital component in human-machine interaction. Dimensional models of affect such as those using valence and arousal have advantages over traditional categorical ones due to the complexity of emotional states in humans. However, dimensional emotion annotations are difficult and expensive to collect, therefore they are not as prevalent in the affective computing community. To address these issues, we propose a method to generate synthetic images from existing categorical emotion datasets using face morphing as well as dimensional labels in the circumplex space with full control over the resulting sample distribution, while achieving augmentation factors of at least 20x or more.

2021-09-01

2103.02937

Silvio Barra Dr

Silvio Barra, Carmen Bisogni, Maria De Marsico, Stefano Ricciardi

Visual Question Answering: which investigated applications?

Pattern Recognition Letters 2021

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Visual Question Answering (VQA) is an extremely stimulating and challenging research area where Computer Vision (CV) and Natural Language Processig (NLP) have recently met. In image captioning and video summarization, the semantic information is completely contained in still images or video dynamics, and it has only to be mined and expressed in a human-consistent way. Differently from this, in VQA semantic information in the same media must be compared with the semantics implied by a question expressed in natural language, doubling the artificial intelligence-related effort. Some recent surveys about VQA approaches have focused on methods underlying either the image-related processing or the verbal-related one, or on the way to consistently fuse the conveyed information. Possible applications are only suggested, and, in fact, most cited works rely on general-purpose datasets that are used to assess the building blocks of a VQA system. This paper rather considers the proposals that focus on real-world applications, possibly using as benchmarks suitable data bound to the application domain. The paper also reports about some recent challenges in VQA research.

2021-03-09

2103.02940

Dmitry V. Dylov

Aleksandr Belov and Joel Stadelmann and Sergey Kastryulin and Dmitry V. Dylov

Towards Ultrafast MRI via Extreme k-Space Undersampling and Superresolution

Main text: 10 pages and 8 figures. 18 pages and 14 figures total (Supplementary material included)

MICCAI 2021. Lecture Notes in Computer Science, vol 12906, pp 254-264

10.1007/978-3-030-87231-1_25

cs.CV eess.IV

http://creativecommons.org/licenses/by-nc-nd/4.0/

We went below the MRI acceleration factors (a.k.a., k-space undersampling) reported by all published papers that reference the original fastMRI challenge, and then considered powerful deep learning based image enhancement methods to compensate for the underresolved images. We thoroughly study the influence of the sampling patterns, the undersampling and the downscaling factors, as well as the recovery models on the final image quality for both the brain and the knee fastMRI benchmarks. The quality of the reconstructed images surpasses that of the other methods, yielding an MSE of 0.00114, a PSNR of 29.6 dB, and an SSIM of 0.956 at x16 acceleration factor. More extreme undersampling factors of x32 and x64 are also investigated, holding promise for certain clinical applications such as computer-assisted surgery or radiation planning. We survey 5 expert radiologists to assess 100 pairs of images and show that the recovered undersampled images statistically preserve their diagnostic value.

2021-10-01

2103.02943

Jose Maria Font

Jose M. Font and Tobias Mahlmann

The Dota 2 Bot Competition

6 pages

IEEE Transactions on Games 2018

10.1109/TG.2018.2834566

cs.AI

http://creativecommons.org/licenses/by/4.0/

Multiplayer Online Battle Area (MOBA) games are a recent huge success both in the video game industry and the international eSports scene. These games encourage team coordination and cooperation, short and long-term planning, within a real-time combined action and strategy gameplay. Artificial Intelligence and Computational Intelligence in Games research competitions offer a wide variety of challenges regarding the study and application of AI techniques to different game genres. These events are widely accepted by the AI/CI community as a sort of AI benchmarking that strongly influences many other research areas in the field. This paper presents and describes in detail the Dota 2 Bot competition and the Dota 2 AI framework that supports it. This challenge aims to join both, MOBAs and AI/CI game competitions, inviting participants to submit AI controllers for the successful MOBA \textit{Defense of the Ancients 2} (Dota 2) to play in 1v1 matches, which aims for fostering research on AI techniques for real-time games. The Dota 2 AI framework makes use of the actual Dota 2 game modding capabilities to enable to connect external AI controllers to actual Dota 2 game matches using the original Free-to-Play game.se of the actual Dota 2 game modding capabilities to enable to connect external AI controllers to actual Dota 2 game matches using the original Free-to-Play game.

2021-03-05

2103.03113

Wei Huang

Wei Huang, Yayong Li, Weitao Du, Jie Yin, Richard Yi Da Xu, Ling Chen, and Miao Zhang

Towards Deepening Graph Neural Networks: A GNTK-based Optimization Perspective

26 pages

ICLR 2022

cs.LG cs.AI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Graph convolutional networks (GCNs) and their variants have achieved great success in dealing with graph-structured data. Nevertheless, it is well known that deep GCNs suffer from the over-smoothing problem, where node representations tend to be indistinguishable as more layers are stacked up. The theoretical research to date on deep GCNs has focused primarily on expressive power rather than trainability, an optimization perspective. Compared to expressivity, trainability attempts to address a more fundamental question: Given a sufficiently expressive space of models, can we successfully find a good solution via gradient descent-based optimizers? This work fills this gap by exploiting the Graph Neural Tangent Kernel (GNTK), which governs the optimization trajectory under gradient descent for wide GCNs. We formulate the asymptotic behaviors of GNTK in the large depth, which enables us to reveal the dropping trainability of wide and deep GCNs at an exponential rate in the optimization process. Additionally, we extend our theoretical framework to analyze residual connection-based techniques, which are found to be merely able to mitigate the exponential decay of trainability mildly. Inspired by our theoretical insights on trainability, we propose Critical DropEdge, a connectivity-aware and graph-adaptive sampling method, to alleviate the exponential decay problem more fundamentally. Experimental evaluation consistently confirms using our proposed method can achieve better results compared to relevant counterparts with both infinite-width and finite-width.

2022-04-22

2103.03133

\v{S}imon Bil\'ik

Simon Bilik, Lukas Kratochvila, Adam Ligocki, Ondrej Bostik, Tomas Zemcik, Matous Hybl, Karel Horak, Ludek Zalud

Visual diagnosis of the Varroa destructor parasitic mite in honeybees using object detector techniques

Sensors, 21-8 (2021), 2764-2780

10.3390/s21082764

BUT171160

cs.CV cs.AI cs.LG

http://creativecommons.org/licenses/by/4.0/

The Varroa destructor mite is one of the most dangerous Honey Bee (Apis mellifera) parasites worldwide and the bee colonies have to be regularly monitored in order to control its spread. Here we present an object detector based method for health state monitoring of bee colonies. This method has the potential for online measurement and processing. In our experiment, we compare the YOLO and SSD object detectors along with the Deep SVDD anomaly detector. Based on the custom dataset with 600 ground-truth images of healthy and infected bees in various scenes, the detectors reached a high F1 score up to 0.874 in the infected bee detection and up to 0.727 in the detection of the Varroa Destructor mite itself. The results demonstrate the potential of this approach, which will be later used in the real-time computer vision based honey bee inspection system. To the best of our knowledge, this study is the first one using object detectors for this purpose. We expect that performance of those object detectors will enable us to inspect the health status of the honey bee colonies.

2023-05-01

2103.03231

Thomas Neff

Thomas Neff, Pascal Stadlbauer, Mathias Parger, Andreas Kurz, Joerg H. Mueller, Chakravarty R. Alla Chaitanya, Anton Kaplanyan, Markus Steinberger

DONeRF: Towards Real-Time Rendering of Compact Neural Radiance Fields using Depth Oracle Networks

Accepted to EGSR 2021 in the CGF track; Project website: https://depthoraclenerf.github.io/

Computer Graphics Forum Volume 40, Issue 4, 2021

10.1111/cgf.14340

cs.CV cs.GR

http://creativecommons.org/licenses/by/4.0/

The recent research explosion around implicit neural representations, such as NeRF, shows that there is immense potential for implicitly storing high-quality scene and lighting information in compact neural networks. However, one major limitation preventing the use of NeRF in real-time rendering applications is the prohibitive computational cost of excessive network evaluations along each view ray, requiring dozens of petaFLOPS. In this work, we bring compact neural representations closer to practical rendering of synthetic content in real-time applications, such as games and virtual reality. We show that the number of samples required for each view ray can be significantly reduced when samples are placed around surfaces in the scene without compromising image quality. To this end, we propose a depth oracle network that predicts ray sample locations for each view ray with a single network evaluation. We show that using a classification network around logarithmically discretized and spherically warped depth values is essential to encode surface locations rather than directly estimating depth. The combination of these techniques leads to DONeRF, our compact dual network design with a depth oracle network as its first step and a locally sampled shading network for ray accumulation. With DONeRF, we reduce the inference costs by up to 48x compared to NeRF when conditioning on available ground truth depth information. Compared to concurrent acceleration methods for raymarching-based neural representations, DONeRF does not require additional memory for explicit caching or acceleration structures, and can render interactively (20 frames per second) on a single GPU.

2021-06-30

2103.03305

Kevin Xu

Mohammadreza Nemati, Haonan Zhang, Michael Sloma, Dulat Bekbolsynov, Hong Wang, Stanislaw Stepkowski, and Kevin S. Xu

Predicting Kidney Transplant Survival using Multiple Feature Representations for HLAs

Extended version of AIME 2021 conference paper

Proceedings of the 19th International Conference on Artificial Intelligence in Medicine (2021) 51-60

cs.LG cs.AI stat.AP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Kidney transplantation can significantly enhance living standards for people suffering from end-stage renal disease. A significant factor that affects graft survival time (the time until the transplant fails and the patient requires another transplant) for kidney transplantation is the compatibility of the Human Leukocyte Antigens (HLAs) between the donor and recipient. In this paper, we propose 4 new biologically-relevant feature representations for incorporating HLA information into machine learning-based survival analysis algorithms. We evaluate our proposed HLA feature representations on a database of over 100,000 transplants and find that they improve prediction accuracy by about 1%, modest at the patient level but potentially significant at a societal level. Accurate prediction of survival times can improve transplant survival outcomes, enabling better allocation of donors to recipients and reducing the number of re-transplants due to graft failure with poorly matched donors.

2022-07-07

2103.03328

Aleksandar Vakanski

Aleksandar Vakanski, Min Xian

Evaluation of Complexity Measures for Deep Learning Generalization in Medical Image Analysis

15 pages, 4 figures

IEEE International Workshop on Machine Learning and Signal Processing (MLSP 2021), Gold Coast, Australia, pp. 1-6, 2021

10.1109/MLSP52302.2021.9596501

cs.CV cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The generalization performance of deep learning models for medical image analysis often decreases on images collected with different devices for data acquisition, device settings, or patient population. A better understanding of the generalization capacity on new images is crucial for clinicians' trustworthiness in deep learning. Although significant research efforts have been recently directed toward establishing generalization bounds and complexity measures, still, there is often a significant discrepancy between the predicted and actual generalization performance. As well, related large empirical studies have been primarily based on validation with general-purpose image datasets. This paper presents an empirical study that investigates the correlation between 25 complexity measures and the generalization abilities of supervised deep learning classifiers for breast ultrasound images. The results indicate that PAC-Bayes flatness-based and path norm-based measures produce the most consistent explanation for the combination of models and data. We also investigate the use of multi-task classification and segmentation approach for breast images, and report that such learning approach acts as an implicit regularizer and is conducive toward improved generalization.

2023-07-20

2103.03335

Leonid Boytsov

Iurii Mokrii, Leonid Boytsov, Pavel Braslavski

A Systematic Evaluation of Transfer Learning and Pseudo-labeling with BERT-based Ranking Models

SIGIR 2021 (44th International ACM SIGIR Conference on Research and Development in Information Retrieval)

10.1145/3404835.3463093

cs.IR cs.CL

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Due to high annotation costs making the best use of existing human-created training data is an important research direction. We, therefore, carry out a systematic evaluation of transferability of BERT-based neural ranking models across five English datasets. Previous studies focused primarily on zero-shot and few-shot transfer from a large dataset to a dataset with a small number of queries. In contrast, each of our collections has a substantial number of queries, which enables a full-shot evaluation mode and improves reliability of our results. Furthermore, since source datasets licences often prohibit commercial use, we compare transfer learning to training on pseudo-labels generated by a BM25 scorer. We find that training on pseudo-labels -- possibly with subsequent fine-tuning using a modest number of annotated queries -- can produce a competitive or better model compared to transfer learning. Yet, it is necessary to improve the stability and/or effectiveness of the few-shot training, which, sometimes, can degrade performance of a pretrained model.

2021-11-23

2103.03359

Amol Kelkar

Amol Kelkar

Cognitive Homeostatic Agents

Accepted at AAMAS2021 Blue Sky Ideas Track

In Proc. of the 20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021), Online, May 3-7, 2021, IFAAMAS, 5 pages

10.5555/3461017.3461021

cs.AI cs.LG cs.MA cs.NE

http://creativecommons.org/licenses/by/4.0/

Human brain has been used as an inspiration for building autonomous agents, but it is not obvious what level of computational description of the brain one should use. This has led to overly opinionated symbolic approaches and overly unstructured connectionist approaches. We propose that using homeostasis as the computational description provides a good compromise. Similar to how physiological homeostasis is the regulation of certain homeostatic variables, cognition can be interpreted as the regulation of certain 'cognitive homeostatic variables'. We present an outline of a Cognitive Homeostatic Agent, built as a hierarchy of physiological and cognitive homeostatic subsystems and describe structures and processes to guide future exploration. We expect this to be a fruitful line of investigation towards building sophisticated artificial agents that can act flexibly in complex environments, and produce behaviors indicating planning, thinking and feelings.

2021-05-04

2103.03413

Yi-Lin Tsai

Yi-Lin Tsai (1), Chetanya Rastogi (2), Peter K. Kitanidis (1, 3, and 4), Christopher B. Field (3, 5, and 6) ((1) Department of Civil and Environmental Engineering, Stanford University, Stanford, CA, USA, (2) Department of Computer Science, Stanford University, Stanford, CA, USA, (3) Woods Institute for the Environment, Stanford University, Stanford, CA, USA, (4) Institute for Computational and Mathematical Engineering, Stanford University, Stanford, CA, USA, (5) Department of Biology, Stanford University, Stanford, CA, USA, (6) Department of Earth System Science, Stanford University, Stanford, CA, USA)

Routing algorithms as tools for integrating social distancing with emergency evacuation

Sci Rep 11, 19623 (2021)

10.1038/s41598-021-98643-z

cs.AI cs.CY cs.HC cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

One of the lessons from the COVID-19 pandemic is the importance of social distancing, even in challenging circumstances such as pre-hurricane evacuation. To explore the implications of integrating social distancing with evacuation operations, we describe this evacuation process as a Capacitated Vehicle Routing Problem (CVRP) and solve it using a DNN (Deep Neural Network)-based solution (Deep Reinforcement Learning) and a non-DNN solution (Sweep Algorithm). A central question is whether Deep Reinforcement Learning provides sufficient extra routing efficiency to accommodate increased social distancing in a time-constrained evacuation operation. We found that, in comparison to the Sweep Algorithm, Deep Reinforcement Learning can provide decision-makers with more efficient routing. However, the evacuation time saved by Deep Reinforcement Learning does not come close to compensating for the extra time required for social distancing, and its advantage disappears as the emergency vehicle capacity approaches the number of people per household.

2021-10-15

2103.03438

Tao Zhang

Mengting Xu, Tao Zhang, Zhongnian Li, Mingxia Liu, Daoqiang Zhang

Towards Evaluating the Robustness of Deep Diagnostic Models by Adversarial Attack

This version was accepted in the journal Medical Image Analysis (MedIA)

Medical Image Analysis 69 (2021): 101977

10.1016/j.media.2021.101977

cs.CV cs.AI cs.LG

http://creativecommons.org/licenses/by/4.0/

Deep learning models (with neural networks) have been widely used in challenging tasks such as computer-aided disease diagnosis based on medical images. Recent studies have shown deep diagnostic models may not be robust in the inference process and may pose severe security concerns in clinical practice. Among all the factors that make the model not robust, the most serious one is adversarial examples. The so-called "adversarial example" is a well-designed perturbation that is not easily perceived by humans but results in a false output of deep diagnostic models with high confidence. In this paper, we evaluate the robustness of deep diagnostic models by adversarial attack. Specifically, we have performed two types of adversarial attacks to three deep diagnostic models in both single-label and multi-label classification tasks, and found that these models are not reliable when attacked by adversarial example. We have further explored how adversarial examples attack the models, by analyzing their quantitative classification results, intermediate features, discriminability of features and correlation of estimated labels for both original/clean images and those adversarial ones. We have also designed two new defense methods to handle adversarial examples in deep diagnostic models, i.e., Multi-Perturbations Adversarial Training (MPAdvT) and Misclassification-Aware Adversarial Training (MAAdvT). The experimental results have shown that the use of defense methods can significantly improve the robustness of deep diagnostic models against adversarial attacks.

2021-03-08

2103.03446

Jialong Tang

Jinsong Su, Jialong Tang, Hui Jiang, Ziyao Lu, Yubin Ge, Linfeng Song, Deyi Xiong, Le Sun, Jiebo Luo

Enhanced Aspect-Based Sentiment Analysis Models with Progressive Self-supervised Attention Learning

31 pages. arXiv admin note: text overlap with arXiv:1906.01213

Artificial Intelligence 2021

cs.CL

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In aspect-based sentiment analysis (ABSA), many neural models are equipped with an attention mechanism to quantify the contribution of each context word to sentiment prediction. However, such a mechanism suffers from one drawback: only a few frequent words with sentiment polarities are tended to be taken into consideration for final sentiment decision while abundant infrequent sentiment words are ignored by models. To deal with this issue, we propose a progressive self-supervised attention learning approach for attentional ABSA models. In this approach, we iteratively perform sentiment prediction on all training instances, and continually learn useful attention supervision information in the meantime. During training, at each iteration, context words with the highest impact on sentiment prediction, identified based on their attention weights or gradients, are extracted as words with active/misleading influence on the correct/incorrect prediction for each instance. Words extracted in this way are masked for subsequent iterations. To exploit these extracted words for refining ABSA models, we augment the conventional training objective with a regularization term that encourages ABSA models to not only take full advantage of the extracted active context words but also decrease the weights of those misleading words. We integrate the proposed approach into three state-of-the-art neural ABSA models. Experiment results and in-depth analyses show that our approach yields better attention results and significantly enhances the performance of all three models. We release the source code and trained models at https://github.com/DeepLearnXMU/PSSAttention.

2021-03-08

2103.03448

Jialong Tang

Jialong Tang, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun, Xinyan Xiao, Hua Wu

Syntactic and Semantic-driven Learning for Open Information Extraction

11 pages

Findings of ACL: EMNLP 2020

cs.CL

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

One of the biggest bottlenecks in building accurate, high coverage neural open IE systems is the need for large labelled corpora. The diversity of open domain corpora and the variety of natural language expressions further exacerbate this problem. In this paper, we propose a syntactic and semantic-driven learning approach, which can learn neural open IE models without any human-labelled data by leveraging syntactic and semantic knowledge as noisier, higher-level supervisions. Specifically, we first employ syntactic patterns as data labelling functions and pretrain a base model using the generated labels. Then we propose a syntactic and semantic-driven reinforcement learning algorithm, which can effectively generalize the base model to open situations with high accuracy. Experimental results show that our approach significantly outperforms the supervised counterparts, and can even achieve competitive performance to supervised state-of-the-art (SoA) model

2021-03-08

2103.03460

Hui Tang

Hui Tang and Kui Jia

Vicinal and categorical domain adaptation

Accepted by Pattern Recognition

Pattern Recognition, Volume 115, July 2021, 107907

10.1016/j.patcog.2021.107907

cs.CV stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Unsupervised domain adaptation aims to learn a task classifier that performs well on the unlabeled target domain, by utilizing the labeled source domain. Inspiring results have been acquired by learning domain-invariant deep features via domain-adversarial training. However, its parallel design of task and domain classifiers limits the ability to achieve a finer category-level domain alignment. To promote categorical domain adaptation (CatDA), based on a joint category-domain classifier, we propose novel losses of adversarial training at both domain and category levels. Since the joint classifier can be regarded as a concatenation of individual task classifiers respectively for the two domains, our design principle is to enforce consistency of category predictions between the two task classifiers. Moreover, we propose a concept of vicinal domains whose instances are produced by a convex combination of pairs of instances respectively from the two domains. Intuitively, alignment of the possibly infinite number of vicinal domains enhances that of original domains. We propose novel adversarial losses for vicinal domain adaptation (VicDA) based on CatDA, leading to Vicinal and Categorical Domain Adaptation (ViCatDA). We also propose Target Discriminative Structure Recovery (TDSR) to recover the intrinsic target discrimination damaged by adversarial feature alignment. We also analyze the principles underlying the ability of our key designs to align the joint distributions. Extensive experiments on several benchmark datasets demonstrate that we achieve the new state of the art.

2021-03-08

2103.03483

Md Mohaimenuzzaman

Md Mohaimenuzzaman, Christoph Bergmeir, Ian Thomas West and Bernd Meyer

Environmental Sound Classification on the Edge: A Pipeline for Deep Acoustic Networks on Extremely Resource-Constrained Devices

Pattern Recognition, p.109025 (2022)

10.1016/j.patcog.2022.109025

cs.SD cs.CV cs.LG eess.AS

http://creativecommons.org/licenses/by/4.0/

Significant efforts are being invested to bring state-of-the-art classification and recognition to edge devices with extreme resource constraints (memory, speed, and lack of GPU support). Here, we demonstrate the first deep network for acoustic recognition that is small, flexible and compression-friendly yet achieves state-of-the-art performance for raw audio classification. Rather than handcrafting a once-off solution, we present a generic pipeline that automatically converts a large deep convolutional network via compression and quantization into a network for resource-impoverished edge devices. After introducing ACDNet, which produces above state-of-the-art accuracy on ESC-10 (96.65%), ESC-50 (87.10%), UrbanSound8K (84.45%) and AudioEvent (92.57%), we describe the compression pipeline and show that it allows us to achieve 97.22% size reduction and 97.28% FLOP reduction while maintaining close to state-of-the-art accuracy 96.25%, 83.65%, 78.27% and 89.69% on these datasets. We describe a successful implementation on a standard off-the-shelf microcontroller and, beyond laboratory benchmarks, report successful tests on real-world datasets.

2022-09-21

2103.03509

Seongsik Park

Seongsik Park and Harksoo Kim

Dual Pointer Network for Fast Extraction of Multiple Relations in a Sentence

Applied Sciences (SI: Natural Language Processing: Emerging Neural Approaches and Applications), Vol.10(11), 2020

10.3390/app10113851

cs.CL

http://creativecommons.org/licenses/by-sa/4.0/

Relation extraction is a type of information extraction task that recognizes semantic relationships between entities in a sentence. Many previous studies have focused on extracting only one semantic relation between two entities in a single sentence. However, multiple entities in a sentence are associated through various relations. To address this issue, we propose a relation extraction model based on a dual pointer network with a multi-head attention mechanism. The proposed model finds n-to-1 subject-object relations using a forward object decoder. Then, it finds 1-to-n subject-object relations using a backward subject decoder. Our experiments confirmed that the proposed model outperformed previous models, with an F1-score of 80.8% for the ACE-2005 corpus and an F1-score of 78.3% for the NYT corpus.

2021-03-08

2103.03518

Julen Balzategui

Julen Balzategui, Luka Eciolaza, and Daniel Maestro-Watson

Anomaly detection and automatic labeling for solar cell quality inspection based on Generative Adversarial Network

20 pages, 10 figures, 6 tables. This article is part of the special issue "Condition Monitoring, Field Inspection and Fault Diagnostic Methods for Photovoltaic Systems" Published in MDPI - Sensors: see https://www.mdpi.com/journal/sensors/special_issues/Condition_Monitoring_Field_Inspection_and_Fault_Diagnostic_Methods_for_Photovoltaic_Systems

Sensors 2021, volume 21, issue 13, article-number 4361

10.3390/s21134361

cs.CV cs.LG eess.IV

http://creativecommons.org/licenses/by/4.0/

Quality inspection applications in industry are required to move towards a zero-defect manufacturing scenario, withnon-destructive inspection and traceability of 100 % of produced parts. Developing robust fault detection and classification modelsfrom the start-up of the lines is challenging due to the difficulty in getting enough representative samples of the faulty patternsand the need to manually label them. This work presents a methodology to develop a robust inspection system, targeting thesepeculiarities, in the context of solar cell manufacturing. The methodology is divided into two phases: In the first phase, an anomalydetection model based on a Generative Adversarial Network (GAN) is employed. This model enables the detection and localizationof anomalous patterns within the solar cells from the beginning, using only non-defective samples for training and without anymanual labeling involved. In a second stage, as defective samples arise, the detected anomalies will be used as automaticallygenerated annotations for the supervised training of a Fully Convolutional Network that is capable of detecting multiple types offaults. The experimental results using 1873 EL images of monocrystalline cells show that (a) the anomaly detection scheme can beused to start detecting features with very little available data, (b) the anomaly detection may serve as automatic labeling in order totrain a supervised model, and (c) segmentation and classification results of supervised models trained with automatic labels arecomparable to the ones obtained from the models trained with manual labels.

2021-07-08

2103.03638

Mark Niklas M\"uller

Mark Niklas M\"uller, Gleb Makarchuk, Gagandeep Singh, Markus P\"uschel, Martin Vechev

PRIMA: General and Precise Neural Network Certification via Scalable Convex Hull Approximations

29 pages, 18 figures, 6 tables

Proceedings of the ACM on Programming Languages, Volume 6, Issue POPL, January 2022, Article No.: 43, pp 1-33

10.1145/3498704

cs.AI cs.LG

http://creativecommons.org/licenses/by/4.0/

Formal verification of neural networks is critical for their safe adoption in real-world applications. However, designing a precise and scalable verifier which can handle different activation functions, realistic network architectures and relevant specifications remains an open and difficult challenge. In this paper, we take a major step forward in addressing this challenge and present a new verification framework, called PRIMA. PRIMA is both (i) general: it handles any non-linear activation function, and (ii) precise: it computes precise convex abstractions involving multiple neurons via novel convex hull approximation algorithms that leverage concepts from computational geometry. The algorithms have polynomial complexity, yield fewer constraints, and minimize precision loss. We evaluate the effectiveness of PRIMA on a variety of challenging tasks from prior work. Our results show that PRIMA is significantly more precise than the state-of-the-art, verifying robustness to input perturbations for up to 20%, 30%, and 34% more images than existing work on ReLU-, Sigmoid-, and Tanh-based networks, respectively. Further, PRIMA enables, for the first time, the precise verification of a realistic neural network for autonomous driving within a few minutes.

2022-03-01

2103.03653

Maciej Besta

Maciej Besta, Zur Vonarburg-Shmaria, Yannick Schaffner, Leonardo Schwarz, Grzegorz Kwasniewski, Lukas Gianinazzi, Jakub Beranek, Kacper Janda, Tobias Holenstein, Sebastian Leisinger, Peter Tatkowski, Esref Ozdemir, Adrian Balla, Marcin Copik, Philipp Lindenberger, Pavel Kalvoda, Marek Konieczny, Onur Mutlu, Torsten Hoefler

GraphMineSuite: Enabling High-Performance and Programmable Graph Mining Algorithms with Set Algebra

International Conference on Very Large Data Bases (VLDB), 2021

cs.DC cs.CV cs.DS cs.MS cs.PF

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We propose GraphMineSuite (GMS): the first benchmarking suite for graph mining that facilitates evaluating and constructing high-performance graph mining algorithms. First, GMS comes with a benchmark specification based on extensive literature review, prescribing representative problems, algorithms, and datasets. Second, GMS offers a carefully designed software platform for seamless testing of different fine-grained elements of graph mining algorithms, such as graph representations or algorithm subroutines. The platform includes parallel implementations of more than 40 considered baselines, and it facilitates developing complex and fast mining algorithms. High modularity is possible by harnessing set algebra operations such as set intersection and difference, which enables breaking complex graph mining algorithms into simple building blocks that can be separately experimented with. GMS is supported with a broad concurrency analysis for portability in performance insights, and a novel performance metric to assess the throughput of graph mining algorithms, enabling more insightful evaluation. As use cases, we harness GMS to rapidly redesign and accelerate state-of-the-art baselines of core graph mining problems: degeneracy reordering (by up to >2x), maximal clique listing (by up to >9x), k-clique listing (by 1.1x), and subgraph isomorphism (by up to 2.5x), also obtaining better theoretical performance bounds.

2023-08-01

2103.03703

Tariq Bdair

Tariq Bdair, Nassir Navab, and Shadi Albarqouni

Semi-Supervised Federated Peer Learning for Skin Lesion Classification

Accepted for publication at the Journal of Machine Learning for Biomedical Imaging (MELBA) [https://www.melba-journal.org%E2%80%9D]https://www.melba-journal.org

Journal of Machine Learning for Biomedical Imaging (MELBA) 2022

cs.CV

http://creativecommons.org/licenses/by-nc-nd/4.0/

Globally, Skin carcinoma is among the most lethal diseases. Millions of people are diagnosed with this cancer every year. Sill, early detection can decrease the medication cost and mortality rate substantially. The recent improvement in automated cancer classification using deep learning methods has reached a human-level performance requiring a large amount of annotated data assembled in one location, yet, finding such conditions usually is not feasible. Recently, federated learning (FL) has been proposed to train decentralized models in a privacy-preserved fashion depending on labeled data at the client-side, which is usually not available and costly. To address this, we propose \verb!FedPerl!, a semi-supervised federated learning method. Our method is inspired by peer learning from educational psychology and ensemble averaging from committee machines. FedPerl builds communities based on clients' similarities. Then it encourages communities members to learn from each other to generate more accurate pseudo labels for the unlabeled data. We also proposed the peer anonymization (PA) technique to anonymize clients. As a core component of our method, PA is orthogonal to other methods without additional complexity and reduces the communication cost while enhancing performance. Finally, we propose a dynamic peer-learning policy that controls the learning stream to avoid any degradation in the performance, especially for individual clients. Our experimental setup consists of 71,000 skin lesion images collected from 5 publicly available datasets. We test our method in four different scenarios in SSFL. With few annotated data, FedPerl is on par with a state-of-the-art method in skin lesion classification in the standard setup while outperforming SSFLs and the baselines by 1.8% and 15.8%, respectively. Also, it generalizes better to unseen clients while being less sensitive to noisy ones.

2022-04-13

2103.03796

Ruidong Yan

Ruidong Yan, Rui Jiang, Bin Jia, Jin Huang, and Diange Yang

Hybrid Car-Following Strategy based on Deep Deterministic Policy Gradient and Cooperative Adaptive Cruise Control

9 pages, 11 figures

published online 2021

10.1109/TASE.2021.3100709

cs.AI cs.LG cs.SY eess.SY

http://creativecommons.org/licenses/by-nc-nd/4.0/

Deep deterministic policy gradient (DDPG)-based car-following strategy can break through the constraints of the differential equation model due to the ability of exploration on complex environments. However, the car-following performance of DDPG is usually degraded by unreasonable reward function design, insufficient training, and low sampling efficiency. In order to solve this kind of problem, a hybrid car-following strategy based on DDPG and cooperative adaptive cruise control (CACC) is proposed. First, the car-following process is modeled as the Markov decision process to calculate CACC and DDPG simultaneously at each frame. Given a current state, two actions are obtained from CACC and DDPG, respectively. Then, an optimal action, corresponding to the one offering a larger reward, is chosen as the output of the hybrid strategy. Meanwhile, a rule is designed to ensure that the change rate of acceleration is smaller than the desired value. Therefore, the proposed strategy not only guarantees the basic performance of car-following through CACC but also makes full use of the advantages of exploration on complex environments via DDPG. Finally, simulation results show that the car-following performance of the proposed strategy is improved compared with that of DDPG and CACC.

2022-01-12

2103.03827

Mathieu Labb\'e

Mathieu Labb\'e and Fran\c{c}ois Michaud

Multi-Session Visual SLAM for Illumination Invariant Re-Localization in Indoor Environments

20 pages, 7 figures

M. Labb\'e and F. Michaud, Multi-Session Visual SLAM for Illumination-Invariant Re-Localization in Indoor Environments, in Frontiers in Robotics and AI, vol. 9, 2022

10.3389/frobt.2022.801886

cs.RO cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

For robots navigating using only a camera, illumination changes in indoor environments can cause re-localization failures during autonomous navigation. In this paper, we present a multi-session visual SLAM approach to create a map made of multiple variations of the same locations in different illumination conditions. The multi-session map can then be used at any hour of the day for improved re-localization capability. The approach presented is independent of the visual features used, and this is demonstrated by comparing re-localization performance between multi-session maps created using the RTAB-Map library with SURF, SIFT, BRIEF, BRISK, KAZE, DAISY and SuperPoint visual features. The approach is tested on six mapping and six localization sessions recorded at 30 minute intervals during sunset using a Google Tango phone in a real apartment.

2022-06-30

2103.03877

Aydogan Ozcan

Yijie Zhang, Tairan Liu, Manmohan Singh, Yilin Luo, Yair Rivenson, Kirill V. Larin, and Aydogan Ozcan

Neural network-based image reconstruction in swept-source optical coherence tomography using undersampled spectral data

20 Pages, 7 Figures, 1 Table

Light: Science & Applications (2021)

10.1038/s41377-021-00594-7

eess.IV cs.CV cs.LG physics.optics

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Optical Coherence Tomography (OCT) is a widely used non-invasive biomedical imaging modality that can rapidly provide volumetric images of samples. Here, we present a deep learning-based image reconstruction framework that can generate swept-source OCT (SS-OCT) images using undersampled spectral data, without any spatial aliasing artifacts. This neural network-based image reconstruction does not require any hardware changes to the optical set-up and can be easily integrated with existing swept-source or spectral domain OCT systems to reduce the amount of raw spectral data to be acquired. To show the efficacy of this framework, we trained and blindly tested a deep neural network using mouse embryo samples imaged by an SS-OCT system. Using 2-fold undersampled spectral data (i.e., 640 spectral points per A-line), the trained neural network can blindly reconstruct 512 A-lines in ~6.73 ms using a desktop computer, removing spatial aliasing artifacts due to spectral undersampling, also presenting a very good match to the images of the same samples, reconstructed using the full spectral OCT data (i.e., 1280 spectral points per A-line). We also successfully demonstrate that this framework can be further extended to process 3x undersampled spectral data per A-line, with some performance degradation in the reconstructed image quality compared to 2x spectral undersampling. This deep learning-enabled image reconstruction approach can be broadly used in various forms of spectral domain OCT systems, helping to increase their imaging speed without sacrificing image resolution and signal-to-noise ratio.

2021-07-30

2103.03905

Jason Ramapuram

Jason Ramapuram, Yan Wu, Alexandros Kalousis

Kanerva++: extending The Kanerva Machine with differentiable, locally block allocated latent memory

ICLR 2021

cs.NE cs.AI cs.CV cs.LG stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Episodic and semantic memory are critical components of the human memory model. The theory of complementary learning systems (McClelland et al., 1995) suggests that the compressed representation produced by a serial event (episodic memory) is later restructured to build a more generalized form of reusable knowledge (semantic memory). In this work we develop a new principled Bayesian memory allocation scheme that bridges the gap between episodic and semantic memory via a hierarchical latent variable model. We take inspiration from traditional heap allocation and extend the idea of locally contiguous memory to the Kanerva Machine, enabling a novel differentiable block allocated latent memory. In contrast to the Kanerva Machine, we simplify the process of memory writing by treating it as a fully feed forward deterministic process, relying on the stochasticity of the read key distribution to disperse information within the memory. We demonstrate that this allocation scheme improves performance in memory conditional image generation, resulting in new state-of-the-art conditional likelihood values on binarized MNIST (<=41.58 nats/image) , binarized Omniglot (<=66.24 nats/image), as well as presenting competitive performance on CIFAR10, DMLab Mazes, Celeb-A and ImageNet32x32.

2022-02-08

2103.03975

Nico Lang

Nico Lang, Nikolai Kalischek, John Armston, Konrad Schindler, Ralph Dubayah, Jan Dirk Wegner

Global canopy height regression and uncertainty estimation from GEDI LIDAR waveforms with deep ensembles

Remote Sensing of Environment 268 (2022) 112760

10.1016/j.rse.2021.112760

cs.LG cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

NASA's Global Ecosystem Dynamics Investigation (GEDI) is a key climate mission whose goal is to advance our understanding of the role of forests in the global carbon cycle. While GEDI is the first space-based LIDAR explicitly optimized to measure vertical forest structure predictive of aboveground biomass, the accurate interpretation of this vast amount of waveform data across the broad range of observational and environmental conditions is challenging. Here, we present a novel supervised machine learning approach to interpret GEDI waveforms and regress canopy top height globally. We propose a probabilistic deep learning approach based on an ensemble of deep convolutional neural networks(CNN) to avoid the explicit modelling of unknown effects, such as atmospheric noise. The model learns to extract robust features that generalize to unseen geographical regions and, in addition, yields reliable estimates of predictive uncertainty. Ultimately, the global canopy top height estimates produced by our model have an expected RMSE of 2.7 m with low bias.

2021-11-05

2103.03991

Brendan Tidd

Brendan Tidd, Akansel Cosgun, Jurgen Leitner, and Nicolas Hudson

Passing Through Narrow Gaps with Deep Reinforcement Learning

Submitted to 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

In proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2021

cs.RO cs.AI

http://creativecommons.org/licenses/by/4.0/

The U.S. Defense Advanced Research Projects Agency (DARPA) Subterranean Challenge requires teams of robots to traverse difficult and diverse underground environments. Traversing small gaps is one of the challenging scenarios that robots encounter. Imperfect sensor information makes it difficult for classical navigation methods, where behaviours require significant manual fine tuning. In this paper we present a deep reinforcement learning method for autonomously navigating through small gaps, where contact between the robot and the gap may be required. We first learn a gap behaviour policy to get through small gaps (only centimeters wider than the robot). We then learn a goal-conditioned behaviour selection policy that determines when to activate the gap behaviour policy. We train our policies in simulation and demonstrate their effectiveness with a large tracked robot in simulation and on the real platform. In simulation experiments, our approach achieves 93\% success rate when the gap behaviour is activated manually by an operator, and 63\% with autonomous activation using the behaviour selection policy. In real robot experiments, our approach achieves a success rate of 73\% with manual activation, and 40\% with autonomous behaviour selection. While we show the feasibility of our approach in simulation, the difference in performance between simulated and real world scenarios highlight the difficulty of direct sim-to-real transfer for deep reinforcement learning policies. In both the simulated and real world environments alternative methods were unable to traverse the gap.

2021-11-03

2103.04068

Artjoms Gorpincenko

Artjoms Gorpincenko, Geoffrey French, Peter Knight, Mike Challiss, Michal Mackiewicz

Improving Automated Sonar Video Analysis to Notify About Jellyfish Blooms

IEEE Sensors Journal, 21, 4981-4988 (2021)

10.1109/JSEN.2020.3032031

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Human enterprise often suffers from direct negative effects caused by jellyfish blooms. The investigation of a prior jellyfish monitoring system showed that it was unable to reliably perform in a cross validation setting, i.e. in new underwater environments. In this paper, a number of enhancements are proposed to the part of the system that is responsible for object classification. First, the training set is augmented by adding synthetic data, making the deep learning classifier able to generalise better. Then, the framework is enhanced by employing a new second stage model, which analyzes the outputs of the first network to make the final prediction. Finally, weighted loss and confidence threshold are added to balance out true and false positives. With all the upgrades in place, the system can correctly classify 30.16% (comparing to the initial 11.52%) of all spotted jellyfish, keep the amount of false positives as low as 0.91% (comparing to the initial 2.26%) and operate in real-time within the computational constraints of an autonomous embedded platform.

2021-03-09

2103.04077

Xiaofeng Gao

Xiaofeng Gao, Luyao Yuan, Tianmin Shu, Hongjing Lu, Song-Chun Zhu

Show Me What You Can Do: Capability Calibration on Reachable Workspace for Human-Robot Collaboration

8 pages, 6 figures, IEEE Robotics and Automation Letters (RA-L), 2022

X. Gao, L. Yuan, T. Shu, H. Lu and S. -C. Zhu, "Show Me What You Can Do: Capability Calibration on Reachable Workspace for Human-Robot Collaboration," in IEEE Robotics and Automation Letters, doi: 10.1109/LRA.2022.3144779

10.1109/LRA.2022.3144779

cs.RO cs.AI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Aligning humans' assessment of what a robot can do with its true capability is crucial for establishing a common ground between human and robot partners when they collaborate on a joint task. In this work, we propose an approach to calibrate humans' estimate of a robot's reachable workspace through a small number of demonstrations before collaboration. We develop a novel motion planning method, REMP, which jointly optimizes the physical cost and the expressiveness of robot motion to reveal the robot's reachability to a human observer. Our experiments with human participants demonstrate that a short calibration using REMP can effectively bridge the gap between what a non-expert user thinks a robot can reach and the ground truth. We show that this calibration procedure not only results in better user perception, but also promotes more efficient human-robot collaborations in a subsequent joint task.

2022-01-28

2103.04192

Guy Ohayon

Guy Ohayon, Theo Adrai, Gregory Vaksman, Michael Elad, Peyman Milanfar

High Perceptual Quality Image Denoising with a Posterior Sampling CGAN

2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), Montreal, BC, Canada, 2021, pp. 1805-1813

10.1109/ICCVW54120.2021.00207

cs.CV cs.LG eess.IV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The vast work in Deep Learning (DL) has led to a leap in image denoising research. Most DL solutions for this task have chosen to put their efforts on the denoiser's architecture while maximizing distortion performance. However, distortion driven solutions lead to blurry results with sub-optimal perceptual quality, especially in immoderate noise levels. In this paper we propose a different perspective, aiming to produce sharp and visually pleasing denoised images that are still faithful to their clean sources. Formally, our goal is to achieve high perceptual quality with acceptable distortion. This is attained by a stochastic denoiser that samples from the posterior distribution, trained as a generator in the framework of conditional generative adversarial networks (CGAN). Contrary to distortion-based regularization terms that conflict with perceptual quality, we introduce to the CGAN objective a theoretically founded penalty term that does not force a distortion requirement on individual samples, but rather on their mean. We showcase our proposed method with a novel denoiser architecture that achieves the reformed denoising goal and produces vivid and diverse outcomes in immoderate noise levels.

2024-05-21

2103.04314

Jeremy Straub

Jeremy Straub

Expert System Gradient Descent Style Training: Development of a Defensible Artificial Intelligence Technique

Knowledge Based-Systems (2021)

10.1016/j.knosys.2021.107275

cs.LG cs.AI

http://creativecommons.org/licenses/by-nc-nd/4.0/

Artificial intelligence systems, which are designed with a capability to learn from the data presented to them, are used throughout society. These systems are used to screen loan applicants, make sentencing recommendations for criminal defendants, scan social media posts for disallowed content and more. Because these systems don't assign meaning to their complex learned correlation network, they can learn associations that don't equate to causality, resulting in non-optimal and indefensible decisions being made. In addition to making decisions that are sub-optimal, these systems may create legal liability for their designers and operators by learning correlations that violate anti-discrimination and other laws regarding what factors can be used in different types of decision making. This paper presents the use of a machine learning expert system, which is developed with meaning-assigned nodes (facts) and correlations (rules). Multiple potential implementations are considered and evaluated under different conditions, including different network error and augmentation levels and different training levels. The performance of these systems is compared to random and fully connected networks.

2021-07-05

2103.04318

Patrick Reiser

Patrick Reiser, Andre Eberhard and Pascal Friederich

Implementing graph neural networks with TensorFlow-Keras

Softw. Impacts 2021, 9, 100095

10.1016/j.simpa.2021.100095

cs.LG cs.AI

http://creativecommons.org/licenses/by-nc-nd/4.0/

Graph neural networks are a versatile machine learning architecture that received a lot of attention recently. In this technical report, we present an implementation of convolution and pooling layers for TensorFlow-Keras models, which allows a seamless and flexible integration into standard Keras layers to set up graph models in a functional way. This implies the usage of mini-batches as the first tensor dimension, which can be realized via the new RaggedTensor class of TensorFlow best suited for graphs. We developed the Keras Graph Convolutional Neural Network Python package kgcnn based on TensorFlow-Keras that provides a set of Keras layers for graph networks which focus on a transparent tensor structure passed between layers and an ease-of-use mindset.

2023-10-12

2103.04351

David Hoeller

David Hoeller, Lorenz Wellhausen, Farbod Farshidian, Marco Hutter

Learning a State Representation and Navigation in Cluttered and Dynamic Environments

8 pages, 8 figures, 2 tables

IEEE Robotics and Automation Letters 2021

cs.RO cs.AI cs.CV cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this work, we present a learning-based pipeline to realise local navigation with a quadrupedal robot in cluttered environments with static and dynamic obstacles. Given high-level navigation commands, the robot is able to safely locomote to a target location based on frames from a depth camera without any explicit mapping of the environment. First, the sequence of images and the current trajectory of the camera are fused to form a model of the world using state representation learning. The output of this lightweight module is then directly fed into a target-reaching and obstacle-avoiding policy trained with reinforcement learning. We show that decoupling the pipeline into these components results in a sample efficient policy learning stage that can be fully trained in simulation in just a dozen minutes. The key part is the state representation, which is trained to not only estimate the hidden state of the world in an unsupervised fashion, but also helps bridging the reality gap, enabling successful sim-to-real transfer. In our experiments with the quadrupedal robot ANYmal in simulation and in reality, we show that our system can handle noisy depth images, avoid dynamic obstacles unseen during training, and is endowed with local spatial awareness.

2021-03-09

2103.04364

Zaid Tahir

Zaid Tahir, Rob Alexander

Coverage based testing for V&V and Safety Assurance of Self-driving Autonomous Vehicles: A Systematic Literature Review

IEEE International Conference On Artificial Intelligence Testing (AITest), Oxford, UK, 2020

10.1109/AITEST49225.2020.00011

cs.AI cs.RO cs.SE

http://creativecommons.org/licenses/by/4.0/

Self-driving Autonomous Vehicles (SAVs) are gaining more interest each passing day by the industry as well as the general public. Tech and automobile companies are investing huge amounts of capital in research and development of SAVs to make sure they have a head start in the SAV market in the future. One of the major hurdles in the way of SAVs making it to the public roads is the lack of confidence of public in the safety aspect of SAVs. In order to assure safety and provide confidence to the public in the safety of SAVs, researchers around the world have used coverage-based testing for Verification and Validation (V&V) and safety assurance of SAVs. The objective of this paper is to investigate the coverage criteria proposed and coverage maximizing techniques used by researchers in the last decade up till now, to assure safety of SAVs. We conduct a Systematic Literature Review (SLR) for this investigation in our paper. We present a classification of existing research based on the coverage criteria used. Several research gaps and research directions are also provided in this SLR to enable further research in this domain. This paper provides a body of knowledge in the domain of safety assurance of SAVs. We believe the results of this SLR will be helpful in the progression of V&V and safety assurance of SAVs.

2021-03-09

2103.04384

Coloma Ballester

Patricia Vitoria and Coloma Ballester

Automatic Flare Spot Artifact Detection and Removal in Photographs

Journal of Mathematical Imaging and Vision, 2019

Journal of Mathematical Imaging and Vision, 2019

eess.IV cs.CV

http://creativecommons.org/licenses/by/4.0/

Flare spot is one type of flare artifact caused by a number of conditions, frequently provoked by one or more high-luminance sources within or close to the camera field of view. When light rays coming from a high-luminance source reach the front element of a camera, it can produce intra-reflections within camera elements that emerge at the film plane forming non-image information or flare on the captured image. Even though preventive mechanisms are used, artifacts can appear. In this paper, we propose a robust computational method to automatically detect and remove flare spot artifacts. Our contribution is threefold: firstly, we propose a characterization which is based on intrinsic properties that a flare spot is likely to satisfy; secondly, we define a new confidence measure able to select flare spots among the candidates; and, finally, a method to accurately determine the flare region is given. Then, the detected artifacts are removed by using exemplar-based inpainting. We show that our algorithm achieve top-tier quantitative and qualitative performance.

2021-03-10

2103.04386

Serge Sharoff

Nouran Khallaf, Serge Sharoff

Automatic Difficulty Classification of Arabic Sentences

Accepted at WANLP 2021

The Sixth Arabic Natural Language Processing Workshop (WANLP 2021)

cs.CL

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this paper, we present a Modern Standard Arabic (MSA) Sentence difficulty classifier, which predicts the difficulty of sentences for language learners using either the CEFR proficiency levels or the binary classification as simple or complex. We compare the use of sentence embeddings of different kinds (fastText, mBERT , XLM-R and Arabic-BERT), as well as traditional language features such as POS tags, dependency trees, readability scores and frequency lists for language learners. Our best results have been achieved using fined-tuned Arabic-BERT. The accuracy of our 3-way CEFR classification is F-1 of 0.80 and 0.75 for Arabic-Bert and XLM-R classification respectively and 0.71 Spearman correlation for regression. Our binary difficulty classifier reaches F-1 0.94 and F-1 0.98 for sentence-pair semantic similarity classifier.

2021-03-09

2103.04421

Xin Yuan

Xin Yuan and David J. Brady and Aggelos K. Katsaggelos

Snapshot Compressive Imaging: Principle, Implementation, Theory, Algorithms and Applications

Extension of X. Yuan, D. J. Brady and A. K. Katsaggelos, "Snapshot Compressive Imaging: Theory, Algorithms, and Applications," in IEEE Signal Processing Magazine, vol. 38, no. 2, pp. 65-88, March 2021, doi: 10.1109/MSP.2020.3023869

in IEEE Signal Processing Magazine, vol. 38, no. 2, pp. 65-88, March 2021

10.1109/MSP.2020.3023869.

eess.IV cs.CV

http://creativecommons.org/licenses/by/4.0/

Capturing high-dimensional (HD) data is a long-term challenge in signal processing and related fields. Snapshot compressive imaging (SCI) uses a two-dimensional (2D) detector to capture HD ($\ge3$D) data in a {\em snapshot} measurement. Via novel optical designs, the 2D detector samples the HD data in a {\em compressive} manner; following this, algorithms are employed to reconstruct the desired HD data-cube. SCI has been used in hyperspectral imaging, video, holography, tomography, focal depth imaging, polarization imaging, microscopy, \etc.~Though the hardware has been investigated for more than a decade, the theoretical guarantees have only recently been derived. Inspired by deep learning, various deep neural networks have also been developed to reconstruct the HD data-cube in spectral SCI and video SCI. This article reviews recent advances in SCI hardware, theory and algorithms, including both optimization-based and deep-learning-based algorithms. Diverse applications and the outlook of SCI are also discussed.

2021-03-10

2103.04485

Ming Zhu

Xiao-Yang Liu, Ming Zhu

Convolutional Graph-Tensor Net for Graph Data Completion

IJCAI 2021

cs.LG cs.AI

http://creativecommons.org/licenses/by/4.0/

Graph data completion is a fundamentally important issue as data generally has a graph structure, e.g., social networks, recommendation systems, and the Internet of Things. We consider a graph where each node has a data matrix, represented as a \textit{graph-tensor} by stacking the data matrices in the third dimension. In this paper, we propose a \textit{Convolutional Graph-Tensor Net} (\textit{Conv GT-Net}) for the graph data completion problem, which uses deep neural networks to learn the general transform of graph-tensors. The experimental results on the ego-Facebook data sets show that the proposed \textit{Conv GT-Net} achieves significant improvements on both completion accuracy (50\% higher) and completion speed (3.6x $\sim$ 8.1x faster) over the existing algorithms.

2023-03-03

2103.04493

Qiaojun Feng

Qiaojun Feng, Yue Meng, Mo Shan, Nikolay Atanasov

Localization and Mapping using Instance-specific Mesh Models

8 pages, 9 figures

2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Macau, China, 2019, pp. 4985-4991

10.1109/IROS40897.2019.8967662

cs.CV cs.RO

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper focuses on building semantic maps, containing object poses and shapes, using a monocular camera. This is an important problem because robots need rich understanding of geometry and context if they are to shape the future of transportation, construction, and agriculture. Our contribution is an instance-specific mesh model of object shape that can be optimized online based on semantic information extracted from camera images. Multi-view constraints on the object shape are obtained by detecting objects and extracting category-specific keypoints and segmentation masks. We show that the errors between projections of the mesh model and the observed keypoints and masks can be differentiated in order to obtain accurate instance-specific object shapes. We evaluate the performance of the proposed approach in simulation and on the KITTI dataset by building maps of car poses and shapes.

2021-03-09

2103.04494

Qiaojun Feng

Qiaojun Feng, Nikolay Atanasov

Fully Convolutional Geometric Features for Category-level Object Alignment

7 pages, 9 figures

2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Las Vegas, NV, USA, 2020, pp. 8492-8498

10.1109/IROS45743.2020.9341550

cs.CV cs.RO

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper focuses on pose registration of different object instances from the same category. This is required in online object mapping because object instances detected at test time usually differ from the training instances. Our approach transforms instances of the same category to a normalized canonical coordinate frame and uses metric learning to train fully convolutional geometric features. The resulting model is able to generate pairs of matching points between the instances, allowing category-level registration. Evaluation on both synthetic and real-world data shows that our method provides robust features, leading to accurate alignment of instances with different shapes.

2021-03-09

2103.04526

Pengbo Liu

Pengbo Liu, Li Xiao, S. Kevin Zhou

Incremental Learning for Multi-organ Segmentation with Partially Labeled Datasets

Medical Image Computing and Computer Assisted Intervention--MICCAI 2022: 25th International Conference, Singapore, September 18--22, 2022, Proceedings, Part IV

10.1007/978-3-031-16440-8_68

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

There exists a large number of datasets for organ segmentation, which are partially annotated, and sequentially constructed. A typical dataset is constructed at a certain time by curating medical images and annotating the organs of interest. In other words, new datasets with annotations of new organ categories are built over time. To unleash the potential behind these partially labeled, sequentially-constructed datasets, we propose to learn a multi-organ segmentation model through incremental learning (IL). In each IL stage, we lose access to the previous annotations, whose knowledge is assumingly captured by the current model, and gain the access to a new dataset with annotations of new organ categories, from which we learn to update the organ segmentation model to include the new organs. We give the first attempt to conjecture that the different distribution is the key reason for 'catastrophic forgetting' that commonly exists in IL methods, and verify that IL has the natural adaptability to medical image scenarios. Extensive experiments on five open-sourced datasets are conducted to prove the effectiveness of our method and the conjecture mentioned above.

2023-05-09

2103.04537

Ruizhi Liao

Ruizhi Liao, Daniel Moyer, Miriam Cha, Keegan Quigley, Seth Berkowitz, Steven Horng, Polina Golland, William M. Wells

Multimodal Representation Learning via Maximization of Local Mutual Information

In Proceedings of International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2021

In International Conference on Medical Image Computing and Computer-Assisted Intervention, pp. 273-283. Springer, Cham, 2021

10.1007/978-3-030-87196-3_26

eess.IV cs.CV

http://creativecommons.org/licenses/by/4.0/

We propose and demonstrate a representation learning approach by maximizing the mutual information between local features of images and text. The goal of this approach is to learn useful image representations by taking advantage of the rich information contained in the free text that describes the findings in the image. Our method trains image and text encoders by encouraging the resulting representations to exhibit high local mutual information. We make use of recent advances in mutual information estimation with neural network discriminators. We argue that the sum of local mutual information is typically a lower bound on the global mutual information. Our experimental results in the downstream image classification tasks demonstrate the advantages of using local features for image-text representation learning.

2021-12-16

2103.04555

Zhiwei Qin

Yan Jiao, Xiaocheng Tang, Zhiwei Qin, Shuaiji Li, Fan Zhang, Hongtu Zhu and Jieping Ye

Real-world Ride-hailing Vehicle Repositioning using Deep Reinforcement Learning

Transportation Research: Part C, 2021

cs.LG cs.AI cs.MA

http://creativecommons.org/licenses/by-nc-nd/4.0/

We present a new practical framework based on deep reinforcement learning and decision-time planning for real-world vehicle repositioning on ride-hailing (a type of mobility-on-demand, MoD) platforms. Our approach learns the spatiotemporal state-value function using a batch training algorithm with deep value networks. The optimal repositioning action is generated on-demand through value-based policy search, which combines planning and bootstrapping with the value networks. For the large-fleet problems, we develop several algorithmic features that we incorporate into our framework and that we demonstrate to induce coordination among the algorithmically-guided vehicles. We benchmark our algorithm with baselines in a ride-hailing simulation environment to demonstrate its superiority in improving income efficiency meausred by income-per-hour. We have also designed and run a real-world experiment program with regular drivers on a major ride-hailing platform. We have observed significantly positive results on key metrics comparing our method with experienced drivers who performed idle-time repositioning based on their own expertise.

2021-07-13

2103.04692

Tuomo Hiippala

Tuomo Hiippala and John A. Bateman

Semiotically-grounded distant viewing of diagrams: insights from two multimodal corpora

22 pages, 11 figures. Under review at Digital Scholarship in the Humanities

Digital Scholarship in the Humanities, 2021 (ahead of press)

10.1093/llc/fqab063

cs.CL cs.CV cs.MM

http://creativecommons.org/licenses/by/4.0/

In this article, we bring together theories of multimodal communication and computational methods to study how primary school science diagrams combine multiple expressive resources. We position our work within the field of digital humanities, and show how annotations informed by multimodality research, which target expressive resources and discourse structure, allow imposing structure on the output of computational methods. We illustrate our approach by analysing two multimodal diagram corpora: the first corpus is intended to support research on automatic diagram processing, whereas the second is oriented towards studying diagrams as a mode of communication. Our results show that multimodally-informed annotations can bring out structural patterns in the diagrams, which also extend across diagrams that deal with different topics.

2021-12-23

2103.04736

Rafael Padilha

Rafael Padilha, Tawfiq Salem, Scott Workman, Fernanda A. Andal\'o, Anderson Rocha and Nathan Jacobs

Content-Aware Detection of Temporal Metadata Manipulation

IEEE Transactions on Information Forensics and Security 2022

10.1109/TIFS.2022.3159154

cs.CV cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Most pictures shared online are accompanied by temporal metadata (i.e., the day and time they were taken), which makes it possible to associate an image content with real-world events. Maliciously manipulating this metadata can convey a distorted version of reality. In this work, we present the emerging problem of detecting timestamp manipulation. We propose an end-to-end approach to verify whether the purported time of capture of an outdoor image is consistent with its content and geographic location. We consider manipulations done in the hour and/or month of capture of a photograph. The central idea is the use of supervised consistency verification, in which we predict the probability that the image content, capture time, and geographical location are consistent. We also include a pair of auxiliary tasks, which can be used to explain the network decision. Our approach improves upon previous work on a large benchmark dataset, increasing the classification accuracy from 59.0% to 81.1%. We perform an ablation study that highlights the importance of various components of the method, showing what types of tampering are detectable using our approach. Finally, we demonstrate how the proposed method can be employed to estimate a possible time-of-capture in scenarios in which the timestamp is missing from the metadata.

2022-03-14

2103.04826

Jamal Toutouh

Diego Gabriel Rossit, Jamal Toutouh, and Sergio Nesmachnow

Exact and heuristic approaches for multi-objective garbage accumulation points location in real scenarios

This article has been accepted for publication in the Waste Management journal

Waste Management. 105:467-481 (2020)

10.1016/j.wasman.2020.02.016

cs.OH cs.AI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Municipal solid waste management is a major challenge for nowadays urban societies, because it accounts for a large proportion of public budget and, when mishandled, it can lead to environmental and social problems. This work focuses on the problem of locating waste bins in an urban area, which is considered to have a strong influence in the overall efficiency of the reverse logistic chain. This article contributes with an exact multiobjective approach to solve the waste bin location in which the optimization criteria that are considered are: the accessibility to the system (as quality of service measure), the investment cost, and the required frequency of waste removal from the bins (as a proxy of the posterior routing costs). In this approach, different methods to obtain the objectives ideal and nadir values over the Pareto front are proposed and compared. Then, a family of heuristic methods based on the PageRank algorithm is proposed which aims to optimize the accessibility to the system, the amount of collected waste and the installation cost. The experimental evaluation was performed on real-world scenarios of the cities of Montevideo, Uruguay, and Bah\'ia Blanca, Argentina. The obtained results show the competitiveness of the proposed approaches for constructing a set of candidate solutions that considers the different trade-offs between the optimization criteria.

2021-03-12

2103.04838

Ramanpreet Pahwa Singh

Ramanpreet S Pahwa, Soon Wee Ho, Ren Qin, Richard Chang, Oo Zaw Min, Wang Jie, Vempati Srinivasa Rao, Tin Lay Nwe, Yanjing Yang, Jens Timo Neumann, Ramani Pichumani, Thomas Gregorich

Machine-learning based methodologies for 3d x-ray measurement, characterization and optimization for buried structures in advanced ic packages

7 pages, 9 figures

International Wafer-Level Packaging Conference (IWLPC) 2020

10.23919/IWLPC52010.2020.9375903

cs.CV

http://creativecommons.org/licenses/by-sa/4.0/

For over 40 years lithographic silicon scaling has driven circuit integration and performance improvement in the semiconductor industry. As silicon scaling slows down, the industry is increasingly dependent on IC package technologies to contribute to further circuit integration and performance improvements. This is a paradigm shift and requires the IC package industry to reduce the size and increase the density of internal interconnects on a scale which has never been done before. Traditional package characterization and process optimization relies on destructive techniques such as physical cross-sections and delayering to extract data from internal package features. These destructive techniques are not practical with today's advanced packages. In this paper we will demonstrate how data acquired non-destructively with a 3D X-ray microscope can be enhanced and optimized using machine learning, and can then be used to measure, characterize and optimize the design and production of buried interconnects in advanced IC packages. Test vehicles replicating 2.5D and HBM construction were designed and fabricated, and digital data was extracted from these test vehicles using 3D X-ray and machine learning techniques. The extracted digital data was used to characterize and optimize the design and production of the interconnects and demonstrates a superior alternative to destructive physical analysis. We report an mAP of 0.96 for 3D object detection, a dice score of 0.92 for 3D segmentation, and an average of 2.1um error for 3D metrology on the test dataset. This paper is the first part of a multi-part report.

2021-05-21

2103.04854

Mohammadhossein Bahari

Mohammadhossein Bahari, Ismail Nejjar, Alexandre Alahi

Injecting Knowledge in Data-driven Vehicle Trajectory Predictors

Published in Transportation Research: Part C

Transportation Research Part C: Emerging Technologies, 2021

10.1016/j.trc.2021.103010

cs.AI cs.RO

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Vehicle trajectory prediction tasks have been commonly tackled from two distinct perspectives: either with knowledge-driven methods or more recently with data-driven ones. On the one hand, we can explicitly implement domain-knowledge or physical priors such as anticipating that vehicles will follow the middle of the roads. While this perspective leads to feasible outputs, it has limited performance due to the difficulty to hand-craft complex interactions in urban environments. On the other hand, recent works use data-driven approaches which can learn complex interactions from the data leading to superior performance. However, generalization, \textit{i.e.}, having accurate predictions on unseen data, is an issue leading to unrealistic outputs. In this paper, we propose to learn a "Realistic Residual Block" (RRB), which effectively connects these two perspectives. Our RRB takes any off-the-shelf knowledge-driven model and finds the required residuals to add to the knowledge-aware trajectory. Our proposed method outputs realistic predictions by confining the residual range and taking into account its uncertainty. We also constrain our output with Model Predictive Control (MPC) to satisfy kinematic constraints. Using a publicly available dataset, we show that our method outperforms previous works in terms of accuracy and generalization to new scenes. We will release our code and data split here: https://github.com/vita-epfl/RRB.

2022-03-07

2103.04885

Takuya Kurihana

Takuya Kurihana, Elisabeth Moyer, Rebecca Willett, Davis Gilton, and Ian Foster

Data-driven Cloud Clustering via a Rotationally Invariant Autoencoder

25 pages. Accepted by IEEE Transactions on Geoscience and Remote Sensing (TGRS)

IEEE Transactions on Geoscience and Remote Sensing, 2021

10.1109/TGRS.2021.3098008

cs.CV physics.ao-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Advanced satellite-born remote sensing instruments produce high-resolution multi-spectral data for much of the globe at a daily cadence. These datasets open up the possibility of improved understanding of cloud dynamics and feedback, which remain the biggest source of uncertainty in global climate model projections. As a step towards answering these questions, we describe an automated rotation-invariant cloud clustering (RICC) method that leverages deep learning autoencoder technology to organize cloud imagery within large datasets in an unsupervised fashion, free from assumptions about predefined classes. We describe both the design and implementation of this method and its evaluation, which uses a sequence of testing protocols to determine whether the resulting clusters: (1) are physically reasonable, (i.e., embody scientifically relevant distinctions); (2) capture information on spatial distributions, such as textures; (3) are cohesive and separable in latent space; and (4) are rotationally invariant, (i.e., insensitive to the orientation of an image). Results obtained when these evaluation protocols are applied to RICC outputs suggest that the resultant novel cloud clusters capture meaningful aspects of cloud physics, are appropriately spatially coherent, and are invariant to orientations of input images. Our results support the possibility of using an unsupervised data-driven approach for automated clustering and pattern discovery in cloud imagery.

2021-10-29

2103.04931

Bartosz Sawicki

Maciej \'Swiechowski, Konrad Godlewski, Bartosz Sawicki, Jacek Ma\'ndziuk

Monte Carlo Tree Search: A Review of Recent Modifications and Applications

99 pages, Accepted to Artificial Intelligence Review journal

Artificial Intelligence Review (2023), vol. 56, 2497-2562

10.1007/s10462-022-10228-y

cs.AI cs.LG cs.MA

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Monte Carlo Tree Search (MCTS) is a powerful approach to designing game-playing bots or solving sequential decision problems. The method relies on intelligent tree search that balances exploration and exploitation. MCTS performs random sampling in the form of simulations and stores statistics of actions to make more educated choices in each subsequent iteration. The method has become a state-of-the-art technique for combinatorial games, however, in more complex games (e.g. those with high branching factor or real-time ones), as well as in various practical domains (e.g. transportation, scheduling or security) an efficient MCTS application often requires its problem-dependent modification or integration with other techniques. Such domain-specific modifications and hybrid approaches are the main focus of this survey. The last major MCTS survey has been published in 2012. Contributions that appeared since its release are of particular interest for this review.

2023-04-04

2103.05094

Abdul Waheed

Abdul Waheed, Muskan Goyal, Deepak Gupta, Ashish Khanna, Fadi Al-Turjman, Placido Rogerio Pinheiro

CovidGAN: Data Augmentation Using Auxiliary Classifier GAN for Improved Covid-19 Detection

Accepted at IEEE Access. Received April 30, 2020, accepted May 11, 2020, date of publication May 14, 2020, date of current version May 28, 2020

IEEE Access, vol. 8, pp. 91916-91923, 2020

10.1109/ACCESS.2020.2994762

eess.IV cs.CV cs.LG

http://creativecommons.org/licenses/by/4.0/

Coronavirus (COVID-19) is a viral disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The spread of COVID-19 seems to have a detrimental effect on the global economy and health. A positive chest X-ray of infected patients is a crucial step in the battle against COVID-19. Early results suggest that abnormalities exist in chest X-rays of patients suggestive of COVID-19. This has led to the introduction of a variety of deep learning systems and studies have shown that the accuracy of COVID-19 patient detection through the use of chest X-rays is strongly optimistic. Deep learning networks like convolutional neural networks (CNNs) need a substantial amount of training data. Because the outbreak is recent, it is difficult to gather a significant number of radiographic images in such a short time. Therefore, in this research, we present a method to generate synthetic chest X-ray (CXR) images by developing an Auxiliary Classifier Generative Adversarial Network (ACGAN) based model called CovidGAN. In addition, we demonstrate that the synthetic images produced from CovidGAN can be utilized to enhance the performance of CNN for COVID-19 detection. Classification using CNN alone yielded 85% accuracy. By adding synthetic images produced by CovidGAN, the accuracy increased to 95%. We hope this method will speed up COVID-19 detection and lead to more robust systems of radiology.

2021-03-10

2103.05132

Bonaventure F. P. Dossou

Bonaventure F. P. Dossou and Mohammed Sabry

AfriVEC: Word Embedding Models for African Languages. Case Study of Fon and Nobiin

Africa NLP, EACL 2021

cs.CL cs.AI

http://creativecommons.org/licenses/by/4.0/

From Word2Vec to GloVe, word embedding models have played key roles in the current state-of-the-art results achieved in Natural Language Processing. Designed to give significant and unique vectorized representations of words and entities, those models have proven to efficiently extract similarities and establish relationships reflecting semantic and contextual meaning among words and entities. African Languages, representing more than 31% of the worldwide spoken languages, have recently been subject to lots of research. However, to the best of our knowledge, there are currently very few to none word embedding models for those languages words and entities, and none for the languages under study in this paper. After describing Glove, Word2Vec, and Poincar\'e embeddings functionalities, we build Word2Vec and Poincar\'e word embedding models for Fon and Nobiin, which show promising results. We test the applicability of transfer learning between these models as a landmark for African Languages to jointly involve in mitigating the scarcity of their resources, and attempt to provide linguistic and social interpretations of our results. Our main contribution is to arouse more interest in creating word embedding models proper to African Languages, ready for use, and that can significantly improve the performances of Natural Language Processing downstream tasks on them. The official repository and implementation is at https://github.com/bonaventuredossou/afrivec

2021-03-19

2103.05167

Gihyeon Choi

Gihyeon Choi, Shinhyeok Oh and Harksoo Kim

Improving Document-Level Sentiment Classification Using Importance of Sentences

12 pages, 7 figures, 5 tables

Entropy, Vol.22(12), pp.1-11, 2020.11

10.3390/e22121336

cs.CL

http://creativecommons.org/licenses/by/4.0/

Previous researchers have considered sentiment analysis as a document classification task, in which input documents are classified into predefined sentiment classes. Although there are sentences in a document that support important evidences for sentiment analysis and sentences that do not, they have treated the document as a bag of sentences. In other words, they have not considered the importance of each sentence in the document. To effectively determine polarity of a document, each sentence in the document should be dealt with different degrees of importance. To address this problem, we propose a document-level sentence classification model based on deep neural networks, in which the importance degrees of sentences in documents are automatically determined through gate mechanisms. To verify our new sentiment analysis model, we conducted experiments using the sentiment datasets in the four different domains such as movie reviews, hotel reviews, restaurant reviews, and music reviews. In the experiments, the proposed model outperformed previous state-of-the-art models that do not consider importance differences of sentences in a document. The experimental results show that the importance of sentences should be considered in a document-level sentiment classification task.

2021-03-10

2103.05213

Mingyuan Meng

Mingyuan Meng, Lei Bi, Michael Fulham, David Dagan Feng, and Jinman Kim

Enhancing Medical Image Registration via Appearance Adjustment Networks

Published at NeuroImage

NeuroImage, vol. 259, pp. 119444, 2022

10.1016/j.neuroimage.2022.119444

cs.CV cs.AI eess.IV

http://creativecommons.org/licenses/by-nc-sa/4.0/

Deformable image registration is fundamental for many medical image analyses. A key obstacle for accurate image registration lies in image appearance variations such as the variations in texture, intensities, and noise. These variations are readily apparent in medical images, especially in brain images where registration is frequently used. Recently, deep learning-based registration methods (DLRs), using deep neural networks, have shown computational efficiency that is several orders of magnitude faster than traditional optimization-based registration methods (ORs). DLRs rely on a globally optimized network that is trained with a set of training samples to achieve faster registration. DLRs tend, however, to disregard the target-pair-specific optimization inherent in ORs and thus have degraded adaptability to variations in testing samples. This limitation is severe for registering medical images with large appearance variations, especially since few existing DLRs explicitly take into account appearance variations. In this study, we propose an Appearance Adjustment Network (AAN) to enhance the adaptability of DLRs to appearance variations. Our AAN, when integrated into a DLR, provides appearance transformations to reduce the appearance variations during registration. In addition, we propose an anatomy-constrained loss function through which our AAN generates anatomy-preserving transformations. Our AAN has been purposely designed to be readily inserted into a wide range of DLRs and can be trained cooperatively in an unsupervised and end-to-end manner. We evaluated our AAN with three state-of-the-art DLRs on three well-established public datasets of 3D brain magnetic resonance imaging (MRI). The results show that our AAN consistently improved existing DLRs and outperformed state-of-the-art ORs on registration accuracy, while adding a fractional computational load to existing DLRs.

2022-09-20

2103.05220

Mingyuan Meng

Bingxin Gu, Mingyuan Meng, Lei Bi, Jinman Kim, David Dagan Feng, and Shaoli Song

Prediction of 5-year Progression-Free Survival in Advanced Nasopharyngeal Carcinoma with Pretreatment PET/CT using Multi-Modality Deep Learning-based Radiomics

Accepted at Frontiers in Oncology

Frontiers in Oncology, vol. 12, pp. 899352, 2022

10.3389/fonc.2022.899351

eess.IV cs.CV cs.LG stat.AP

http://creativecommons.org/licenses/by-nc-sa/4.0/

Objective: Deep Learning-based Radiomics (DLR) has achieved great success in medical image analysis and has been considered a replacement for conventional radiomics that relies on handcrafted features. In this study, we aimed to explore the capability of DLR for the prediction of 5-year Progression-Free Survival (PFS) in Nasopharyngeal Carcinoma (NPC) using pretreatment PET/CT. Methods: A total of 257 patients (170/87 in internal/external cohorts) with advanced NPC (TNM stage III or IVa) were enrolled. We developed an end-to-end multi-modality DLR model, in which a 3D convolutional neural network was optimized to extract deep features from pretreatment PET/CT images and predict the probability of 5-year PFS. TNM stage, as a high-level clinical feature, could be integrated into our DLR model to further improve the prognostic performance. To compare conventional radiomics and DLR, 1456 handcrafted features were extracted, and optimal conventional radiomics methods were selected from 54 cross-combinations of 6 feature selection methods and 9 classification methods. In addition, risk group stratification was performed with clinical signature, conventional radiomics signature, and DLR signature. Results: Our multi-modality DLR model using both PET and CT achieved higher prognostic performance than the optimal conventional radiomics method. Furthermore, the multi-modality DLR model outperformed single-modality DLR models using only PET or only CT. For risk group stratification, the conventional radiomics signature and DLR signature enabled significant differences between the high- and low-risk patient groups in both internal and external cohorts, while the clinical signature failed in the external cohort. Conclusion: Our study identified potential prognostic tools for survival prediction in advanced NPC, suggesting that DLR could provide complementary values to the current TNM staging.

2022-09-20

2103.05225

Harel Yedidsion

Harel Yedidsion, Jennifer Suriadinata, Zifan Xu, Stefan Debruyn, Peter Stone

A Scavenger Hunt for Service Robots

6 pages + references + Appendix

the 2021 IEEE International Conference on Robotics and Automation (ICRA), May 30 - June 5, 2021, Xi'an, China

cs.RO cs.AI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Creating robots that can perform general-purpose service tasks in a human-populated environment has been a longstanding grand challenge for AI and Robotics research. One particularly valuable skill that is relevant to a wide variety of tasks is the ability to locate and retrieve objects upon request. This paper models this skill as a Scavenger Hunt (SH) game, which we formulate as a variation of the NP-hard stochastic traveling purchaser problem. In this problem, the goal is to find a set of objects as quickly as possible, given probability distributions of where they may be found. We investigate the performance of several solution algorithms for the SH problem, both in simulation and on a real mobile robot. We use Reinforcement Learning (RL) to train an agent to plan a minimal cost path, and show that the RL agent can outperform a range of heuristic algorithms, achieving near optimal performance. In order to stimulate research on this problem, we introduce a publicly available software stack and associated website that enable users to upload scavenger hunts which robots can download, perform, and learn from to continually improve their performance on future hunts.

2021-03-31

2103.05467

Roni Saputra Permana

Liana Ellen Taylor, Midriem Mirdanies, Roni Permana Saputra

Optimized Object Tracking Technique Using Kalman Filter

10 pages, 14 figures, published in J. Mechatron. Electr. Power Veh. Technol 07 (2016) 57-66

J. Mechatron. Electr. Power Veh. Technol 07 (2016) 57-66

10.14203/j.mev.2016.v7.57-66

cs.CV

http://creativecommons.org/licenses/by/4.0/

This paper focused on the design of an optimized object tracking technique which would minimize the processing time required in the object detection process while maintaining accuracy in detecting the desired moving object in a cluttered scene. A Kalman filter based cropped image is used for the image detection process as the processing time is significantly less to detect the object when a search window is used that is smaller than the entire video frame. This technique was tested with various sizes of the window in the cropping process. MATLAB was used to design and test the proposed method. This paper found that using a cropped image with 2.16 multiplied by the largest dimension of the object resulted in significantly faster processing time while still providing a high success rate of detection and a detected center of the object that was reasonably close to the actual center.

2021-03-10

2103.05481

Damien Pellier

Damien Pellier, Humbert Fiorino

From Classical to Hierarchical: benchmarks for the HTN Track of the International Planning Competition

Proceedings of the International Planning Competition, ICAPS, Nancy, France, 2020

cs.AI

http://creativecommons.org/licenses/by/4.0/

In this short paper, we outline nine classical benchmarks submitted to the first hierarchical planning track of the International Planning competition in 2020. All of these benchmarks are based on the HDDL language. The choice of the benchmarks was based on a questionnaire sent to the HTN community. They are the following: Barman, Childsnack, Rover, Satellite, Blocksworld, Depots, Gripper, and Hiking. In the rest of the paper we give a short description of these benchmarks. All are totally ordered.

2021-03-10

2103.05489

Jan Koh\'ut

Jan Koh\'ut, Michal Hradi\v{s}

TS-Net: OCR Trained to Switch Between Text Transcription Styles

ICDAR 2021: Proceedings, Part IV 16 (pp. 478-493)

10.1007/978-3-030-86337-1_32

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Users of OCR systems, from different institutions and scientific disciplines, prefer and produce different transcription styles. This presents a problem for training of consistent text recognition neural networks on real-world data. We propose to extend existing text recognition networks with a Transcription Style Block (TSB) which can learn from data to switch between multiple transcription styles without any explicit knowledge of transcription rules. TSB is an adaptive instance normalization conditioned by identifiers representing consistently transcribed documents (e.g. single document, documents by a single transcriber, or an institution). We show that TSB is able to learn completely different transcription styles in controlled experiments on artificial data, it improves text recognition accuracy on large-scale real-world data, and it learns semantically meaningful transcription style embedding. We also show how TSB can efficiently adapt to transcription styles of new documents from transcriptions of only a few text lines.

2023-02-14

2103.05529

K. Ruwani Fernando

K. Ruwani M. Fernando and Chris P. Tsokos

Deep and Statistical Learning in Biomedical Imaging: State of the Art in 3D MRI Brain Tumor Segmentation

21 pages, 7 figures

Information Fusion, Volume 92, 2023, Pages 450-465

10.1016/j.inffus.2022.12.013

eess.IV cs.CV

http://creativecommons.org/licenses/by-nc-nd/4.0/

Clinical diagnostic and treatment decisions rely upon the integration of patient-specific data with clinical reasoning. Cancer presents a unique context that influence treatment decisions, given its diverse forms of disease evolution. Biomedical imaging allows noninvasive assessment of disease based on visual evaluations leading to better clinical outcome prediction and therapeutic planning. Early methods of brain cancer characterization predominantly relied upon statistical modeling of neuroimaging data. Driven by the breakthroughs in computer vision, deep learning became the de facto standard in the domain of medical imaging. Integrated statistical and deep learning methods have recently emerged as a new direction in the automation of the medical practice unifying multi-disciplinary knowledge in medicine, statistics, and artificial intelligence. In this study, we critically review major statistical and deep learning models and their applications in brain imaging research with a focus on MRI-based brain tumor segmentation. The results do highlight that model-driven classical statistics and data-driven deep learning is a potent combination for developing automated systems in clinical oncology.

2022-12-23

2103.05564

Marco Pegoraro

Marco Pegoraro, Merih Seran Uysal, Wil M.P. van der Aalst

PROVED: A Tool for Graph Representation and Analysis of Uncertain Event Data

11 pages, 6 figures, 1 table, 16 references

Petri Nets (2021) 476-486

10.1007/978-3-030-76983-3_24

cs.AI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The discipline of process mining aims to study processes in a data-driven manner by analyzing historical process executions, often employing Petri nets. Event data, extracted from information systems (e.g. SAP), serve as the starting point for process mining. Recently, novel types of event data have gathered interest among the process mining community, including uncertain event data. Uncertain events, process traces and logs contain attributes that are characterized by quantified imprecisions, e.g., a set of possible attribute values. The PROVED tool helps to explore, navigate and analyze such uncertain event data by abstracting the uncertain information using behavior graphs and nets, which have Petri nets semantics. Based on these constructs, the tool enables discovery and conformance checking.

2022-04-11

2103.05661

Anca Dragan

Liting Sun, Xiaogang Jia, Anca D. Dragan

On complementing end-to-end human behavior predictors with planning

Robotics: Science and Systems, 2021

cs.AI cs.RO

http://creativecommons.org/licenses/by/4.0/

High capacity end-to-end approaches for human motion (behavior) prediction have the ability to represent subtle nuances in human behavior, but struggle with robustness to out of distribution inputs and tail events. Planning-based prediction, on the other hand, can reliably output decent-but-not-great predictions: it is much more stable in the face of distribution shift (as we verify in this work), but it has high inductive bias, missing important aspects that drive human decisions, and ignoring cognitive biases that make human behavior suboptimal. In this work, we analyze one family of approaches that strive to get the best of both worlds: use the end-to-end predictor on common cases, but do not rely on it for tail events / out-of-distribution inputs -- switch to the planning-based predictor there. We contribute an analysis of different approaches for detecting when to make this switch, using an autonomous driving domain. We find that promising approaches based on ensembling or generative modeling of the training distribution might not be reliable, but that there very simple methods which can perform surprisingly well -- including training a classifier to pick up on tell-tale issues in predicted trajectories.

2021-07-14

2103.05886

Hilmil Pradana

Hilmil Pradana and Keiichi Horio

Tuna Nutriment Tracking using Trajectory Mapping in Application to Aquaculture Fish Tank

2020 Digital Image Computing: Techniques and Applications (DICTA) (2020) 1-8

10.1109/DICTA51227.2020.9363387

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The cost of fish feeding is usually around 40 percent of total production cost. Estimating a state of fishes in a tank and adjusting an amount of nutriments play an important role to manage cost of fish feeding system. Our approach is based on tracking nutriments on videos collected from an active aquaculture fish farm. Tracking approach is applied to acknowledge movement of nutriment to understand more about the fish behavior. Recently, there has been increasing number of researchers focused on developing tracking algorithms to generate more accurate and faster determination of object. Unfortunately, recent studies have shown that efficient and robust tracking of multiple objects with complex relations remain unsolved. Hence, focusing to develop tracking algorithm in aquaculture is more challenging because tracked object has a lot of aquatic variant creatures. By following aforementioned problem, we develop tuna nutriment tracking based on the classical minimum cost problem which consistently performs well in real environment datasets. In evaluation, the proposed method achieved 21.32 pixels and 3.08 pixels for average error distance and standard deviation, respectively. Quantitative evaluation based on the data generated by human annotators shows that the proposed method is valuable for aquaculture fish farm and can be widely applied to real environment datasets.

2021-03-11

2103.05918

Dong Shen

Dong Shen, Shuai Zhao, Jinming Hu, Hao Feng, Deng Cai, Xiaofei He

ES-Net: Erasing Salient Parts to Learn More in Re-Identification

11 pages, 6 figures. Accepted for publication in IEEE Transactions on Image Processing 2021

IEEE Transactions on Image Processing, vol. 30, pp. 1676-1686, 2021

10.1109/TIP.2020.3046904

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

As an instance-level recognition problem, re-identification (re-ID) requires models to capture diverse features. However, with continuous training, re-ID models pay more and more attention to the salient areas. As a result, the model may only focus on few small regions with salient representations and ignore other important information. This phenomenon leads to inferior performance, especially when models are evaluated on small inter-identity variation data. In this paper, we propose a novel network, Erasing-Salient Net (ES-Net), to learn comprehensive features by erasing the salient areas in an image. ES-Net proposes a novel method to locate the salient areas by the confidence of objects and erases them efficiently in a training batch. Meanwhile, to mitigate the over-erasing problem, this paper uses a trainable pooling layer P-pooling that generalizes global max and global average pooling. Experiments are conducted on two specific re-identification tasks (i.e., Person re-ID, Vehicle re-ID). Our ES-Net outperforms state-of-the-art methods on three Person re-ID benchmarks and two Vehicle re-ID benchmarks. Specifically, mAP / Rank-1 rate: 88.6% / 95.7% on Market1501, 78.8% / 89.2% on DuckMTMC-reID, 57.3% / 80.9% on MSMT17, 81.9% / 97.0% on Veri-776, respectively. Rank-1 / Rank-5 rate: 83.6% / 96.9% on VehicleID (Small), 79.9% / 93.5% on VehicleID (Medium), 76.9% / 90.7% on VehicleID (Large), respectively. Moreover, the visualized salient areas show human-interpretable visual explanations for the ranking results.

2021-03-11

2103.05923

XinZhou Dong

Xinzhou Dong, Beihong Jin, Wei Zhuo, Beibei Li, Taofeng Xue

Improving Sequential Recommendation with Attribute-augmented Graph Neural Networks

The 25th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD-2021), May 11-14, 2021, Delhi, India

cs.IR cs.AI

http://creativecommons.org/licenses/by/4.0/

Many practical recommender systems provide item recommendation for different users only via mining user-item interactions but totally ignoring the rich attribute information of items that users interact with. In this paper, we propose an attribute-augmented graph neural network model named Murzim. Murzim takes as input the graphs constructed from the user-item interaction sequences and corresponding item attribute sequences. By combining the GNNs with node aggregation and an attention network, Murzim can capture user preference patterns, generate embeddings for user-item interaction sequences, and then generate recommendations through next-item prediction. We conduct extensive experiments on multiple datasets. Experimental results show that Murzim outperforms several state-of-the-art methods in terms of recall and MRR, which illustrates that Murzim can make use of item attribute information to produce better recommendations. At present, Murzim has been deployed in MX Player, one of India's largest streaming platforms, and is recommending videos for tens of thousands of users.

2021-03-11

2103.05944

Xiuying Chen

Mingfei Guo, Xiuying Chen, Juntao Li, Dongyan Zhao, Rui Yan

How does Truth Evolve into Fake News? An Empirical Study of Fake News Evolution

5 pages, 2 figures

The Web Conference 2021, Workshop on News Recommendation and Intelligence

cs.CL

http://creativecommons.org/licenses/by/4.0/

Automatically identifying fake news from the Internet is a challenging problem in deception detection tasks. Online news is modified constantly during its propagation, e.g., malicious users distort the original truth and make up fake news. However, the continuous evolution process would generate unprecedented fake news and cheat the original model. We present the Fake News Evolution (FNE) dataset: a new dataset tracking the fake news evolution process. Our dataset is composed of 950 paired data, each of which consists of articles representing the three significant phases of the evolution process, which are the truth, the fake news, and the evolved fake news. We observe the features during the evolution and they are the disinformation techniques, text similarity, top 10 keywords, classification accuracy, parts of speech, and sentiment properties.

2021-03-11

2103.05977

Yuan-Gen Wang

Fu-Zhao Ou, Xingyu Chen, Ruixin Zhang, Yuge Huang, Shaoxin Li, Jilin Li, Yong Li, Liujuan Cao, and Yuan-Gen Wang

SDD-FIQA: Unsupervised Face Image Quality Assessment with Similarity Distribution Distance

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021

cs.CV

http://creativecommons.org/licenses/by/4.0/

In recent years, Face Image Quality Assessment (FIQA) has become an indispensable part of the face recognition system to guarantee the stability and reliability of recognition performance in an unconstrained scenario. For this purpose, the FIQA method should consider both the intrinsic property and the recognizability of the face image. Most previous works aim to estimate the sample-wise embedding uncertainty or pair-wise similarity as the quality score, which only considers the information from partial intra-class. However, these methods ignore the valuable information from the inter-class, which is for estimating to the recognizability of face image. In this work, we argue that a high-quality face image should be similar to its intra-class samples and dissimilar to its inter-class samples. Thus, we propose a novel unsupervised FIQA method that incorporates Similarity Distribution Distance for Face Image Quality Assessment (SDD-FIQA). Our method generates quality pseudo-labels by calculating the Wasserstein Distance (WD) between the intra-class similarity distributions and inter-class similarity distributions. With these quality pseudo-labels, we are capable of training a regression network for quality prediction. Extensive experiments on benchmark datasets demonstrate that the proposed SDD-FIQA surpasses the state-of-the-arts by an impressive margin. Meanwhile, our method shows good generalization across different recognition systems.

2021-03-11

2103.06022

Stefan Schrunner

Delmon Arous, Stefan Schrunner, Ingunn Hanson, Nina F.J. Edin, Eirik Malinen

Principal component-based image segmentation: a new approach to outline in vitro cell colonies

Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization (2022)

10.1080/21681163.2022.2035822

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The in vitro clonogenic assay is a technique to study the ability of a cell to form a colony in a culture dish. By optical imaging, dishes with stained colonies can be scanned and assessed digitally. Identification, segmentation and counting of stained colonies play a vital part in high-throughput screening and quantitative assessment of biological assays. Image processing of such pictured/scanned assays can be affected by image/scan acquisition artifacts like background noise and spatially varying illumination, and contaminants in the suspension medium. Although existing approaches tackle these issues, the segmentation quality requires further improvement, particularly on noisy and low contrast images. In this work, we present an objective and versatile machine learning procedure to amend these issues by characterizing, extracting and segmenting inquired colonies using principal component analysis, k-means clustering and a modified watershed segmentation algorithm. The intention is to automatically identify visible colonies through spatial texture assessment and accordingly discriminate them from background in preparation for successive segmentation. The proposed segmentation algorithm yielded a similar quality as manual counting by human observers. High F1 scores (>0.9) and low root-mean-square errors (around 14%) underlined good agreement with ground truth data. Moreover, it outperformed a recent state-of-the-art method. The methodology will be an important tool in future cancer research applications.

2022-02-15

2103.06115

Veronica Sanz

Gabriela Barenboim, Johannes Hirn and Veronica Sanz

Symmetry meets AI

8 pages, 8 figures

SciPost Phys. 11, 014 (2021)

10.21468/SciPostPhys.11.1.014

cs.LG cs.CV hep-ph

http://creativecommons.org/licenses/by/4.0/

We explore whether Neural Networks (NNs) can {\it discover} the presence of symmetries as they learn to perform a task. For this, we train hundreds of NNs on a {\it decoy task} based on well-controlled Physics templates, where no information on symmetry is provided. We use the output from the last hidden layer of all these NNs, projected to fewer dimensions, as the input for a symmetry classification task, and show that information on symmetry had indeed been identified by the original NN without guidance. As an interdisciplinary application of this procedure, we identify the presence and level of symmetry in artistic paintings from different styles such as those of Picasso, Pollock and Van Gogh.

2021-07-21

2103.06123

Hiroshi Yamakawa

Hiroshi Yamakawa

The whole brain architecture approach: Accelerating the development of artificial general intelligence by referring to the brain

28 pages, 10 figures, Preprint submitted to Neural Networks

Neural Networks, Volume 144, December 2021, Pages 478-495

10.1016/j.neunet.2021.09.004

cs.AI cs.LG cs.NE q-bio.NC

http://creativecommons.org/licenses/by-sa/4.0/

The vastness of the design space created by the combination of a large number of computational mechanisms, including machine learning, is an obstacle to creating an artificial general intelligence (AGI). Brain-inspired AGI development, in other words, cutting down the design space to look more like a biological brain, which is an existing model of a general intelligence, is a promising plan for solving this problem. However, it is difficult for an individual to design a software program that corresponds to the entire brain because the neuroscientific data required to understand the architecture of the brain are extensive and complicated. The whole-brain architecture approach divides the brain-inspired AGI development process into the task of designing the brain reference architecture (BRA) -- the flow of information and the diagram of corresponding components -- and the task of developing each component using the BRA. This is called BRA-driven development. Another difficulty lies in the extraction of the operating principles necessary for reproducing the cognitive-behavioral function of the brain from neuroscience data. Therefore, this study proposes the Structure-constrained Interface Decomposition (SCID) method, which is a hypothesis-building method for creating a hypothetical component diagram consistent with neuroscientific findings. The application of this approach has begun for building various regions of the brain. Moving forward, we will examine methods of evaluating the biological plausibility of brain-inspired software. This evaluation will also be used to prioritize different computational mechanisms, which should be merged, associated with the same regions of the brain.

2022-08-16

2103.06168

Tommaso Di Noto

Tommaso Di Noto, Guillaume Marie, Sebastien Tourbier, Yasser Alem\'an-G\'omez, Oscar Esteban, Guillaume Saliou, Meritxell Bach Cuadra, Patric Hagmann, Jonas Richiardi

Towards automated brain aneurysm detection in TOF-MRA: open data, weak labels, and anatomical knowledge

Paper accepted as Original Article in the journal Neuroinformatics (https://link.springer.com/article/10.1007/s12021-022-09597-0)

Neuroinformatics, 2022

10.1007/s12021-022-09597-0

eess.IV cs.CV

http://creativecommons.org/licenses/by-sa/4.0/

Brain aneurysm detection in Time-Of-Flight Magnetic Resonance Angiography (TOF-MRA) has undergone drastic improvements with the advent of Deep Learning (DL). However, performances of supervised DL models heavily rely on the quantity of labeled samples, which are extremely costly to obtain. Here, we present a DL model for aneurysm detection that overcomes the issue with ''weak'' labels: oversized annotations which are considerably faster to create. Our weak labels resulted to be four times faster to generate than their voxel-wise counterparts. In addition, our model leverages prior anatomical knowledge by focusing only on plausible locations for aneurysm occurrence. We frst train and evaluate our model through cross-validation on an in-house TOF-MRA dataset comprising 284 subjects (170 females / 127 healthy controls / 157 patients with 198 aneurysms). On this dataset, our best model achieved a sensitivity of 83%, with False Positive (FP) rate of 0.8 per patient. To assess model generalizability, we then participated in a challenge for aneurysm detection with TOF-MRA data (93 patients, 20 controls, 125 aneurysms). On the public challenge, sensitivity was 68% (FP rate=2.5), ranking 4th/18 on the open leaderboard. We found no signifcant diference in sensitivity between aneurysm risk-of-rupture groups (p=0.75), locations (p=0.72), or sizes (p=0.15). Data, code and model weights are released under permissive licenses. We demonstrate that weak labels and anatomical knowledge can alleviate the necessity for prohibitively expensive voxel-wise annotations.

2022-08-24

2103.06182

Heng Yang

Heng Yang, Chris Doran, Jean-Jacques Slotine

Dynamical Pose Estimation

ICCV 2021 camera ready. Code: https://github.com/hankyang94/DAMP. Video: https://youtu.be/CDYXR1h98Q4

ICCV 2021

cs.CV cs.RO math.DS

http://creativecommons.org/licenses/by/4.0/

We study the problem of aligning two sets of 3D geometric primitives given known correspondences. Our first contribution is to show that this primitive alignment framework unifies five perception problems including point cloud registration, primitive (mesh) registration, category-level 3D registration, absolution pose estimation (APE), and category-level APE. Our second contribution is to propose DynAMical Pose estimation (DAMP), the first general and practical algorithm to solve primitive alignment problem by simulating rigid body dynamics arising from virtual springs and damping, where the springs span the shortest distances between corresponding primitives. We evaluate DAMP in simulated and real datasets across all five problems, and demonstrate (i) DAMP always converges to the globally optimal solution in the first three problems with 3D-3D correspondences; (ii) although DAMP sometimes converges to suboptimal solutions in the last two problems with 2D-3D correspondences, using a scheme for escaping local minima, DAMP always succeeds. Our third contribution is to demystify the surprising empirical performance of DAMP and formally prove a global convergence result in the case of point cloud registration by charactering local stability of the equilibrium points of the underlying dynamical system.

2021-08-13

2103.06205

Florian Kofler

Florian Kofler, Ivan Ezhov, Fabian Isensee, Fabian Balsiger, Christoph Berger, Maximilian Koerner, Beatrice Demiray, Julia Rackerseder, Johannes Paetzold, Hongwei Li, Suprosanna Shit, Richard McKinley, Marie Piraud, Spyridon Bakas, Claus Zimmer, Nassir Navab, Jan Kirschke, Benedikt Wiestler, Bjoern Menze

Are we using appropriate segmentation metrics? Identifying correlates of human expert perception for CNN training beyond rolling the DICE coefficient

Accepted for publication at the Journal of Machine Learning for Biomedical Imaging (MELBA) https://melba-journal.org/2023:002

Machine.Learning.for.Biomedical.Imaging. 2 (2023)

eess.IV cs.CV

http://creativecommons.org/licenses/by/4.0/

Metrics optimized in complex machine learning tasks are often selected in an ad-hoc manner. It is unknown how they align with human expert perception. We explore the correlations between established quantitative segmentation quality metrics and qualitative evaluations by professionally trained human raters. Therefore, we conduct psychophysical experiments for two complex biomedical semantic segmentation problems. We discover that current standard metrics and loss functions correlate only moderately with the segmentation quality assessment of experts. Importantly, this effect is particularly pronounced for clinically relevant structures, such as the enhancing tumor compartment of glioma in brain magnetic resonance and grey matter in ultrasound imaging. It is often unclear how to optimize abstract metrics, such as human expert perception, in convolutional neural network (CNN) training. To cope with this challenge, we propose a novel strategy employing techniques of classical statistics to create complementary compound loss functions to better approximate human expert perception. Across all rating experiments, human experts consistently scored computer-generated segmentations better than the human-curated reference labels. Our results, therefore, strongly question many current practices in medical image segmentation and provide meaningful cues for future research.

2023-05-03

2103.06304

Letitia Parcalabescu

Letitia Parcalabescu, Nils Trost, Anette Frank

What is Multimodality?

Paper accepted for publication at MMSR 2021; 10 pages, 5 figures

Proceedings of the 1st Workshop on Multimodal Semantic Representations (MMSR), 2021, Groningen, Netherlands (Online), Association for Computational Linguistics, p. 1--10

cs.AI cs.CL cs.CV

http://creativecommons.org/licenses/by/4.0/

The last years have shown rapid developments in the field of multimodal machine learning, combining e.g., vision, text or speech. In this position paper we explain how the field uses outdated definitions of multimodality that prove unfit for the machine learning era. We propose a new task-relative definition of (multi)modality in the context of multimodal machine learning that focuses on representations and information that are relevant for a given machine learning task. With our new definition of multimodality we aim to provide a missing foundation for multimodal research, an important component of language grounding and a crucial milestone towards NLU.

2021-08-23

2103.06410

Chenguang Zhu

Chenguang Zhu, Yang Liu, Jie Mei, Michael Zeng

MediaSum: A Large-scale Media Interview Dataset for Dialogue Summarization

Dataset: https://github.com/zcgzcgzcg1/MediaSum/

North American Chapter of the Association for Computational Linguistics (NAACL), Mexico City, Mexico, 2021

cs.CL

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

MediaSum, a large-scale media interview dataset consisting of 463.6K transcripts with abstractive summaries. To create this dataset, we collect interview transcripts from NPR and CNN and employ the overview and topic descriptions as summaries. Compared with existing public corpora for dialogue summarization, our dataset is an order of magnitude larger and contains complex multi-party conversations from multiple domains. We conduct statistical analysis to demonstrate the unique positional bias exhibited in the transcripts of televised and radioed interviews. We also show that MediaSum can be used in transfer learning to improve a model's performance on other dialogue summarization tasks.

2021-03-15

2103.06501

Chi Zhang

Chi Zhang, Zihang Lin, Liheng Xu, Zongliang Li, Wei Tang, Yuehu Liu, Gaofeng Meng, Le Wang, Li Li

Density-aware Haze Image Synthesis by Self-Supervised Content-Style Disentanglement

21 pages, 19 figures, 6 tables

IEEE Transactions on Circuits and Systems for Video Technology

10.1109/TCSVT.2021.3130158

cs.CV cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The key procedure of haze image translation through adversarial training lies in the disentanglement between the feature only involved in haze synthesis, i.e.style feature, and the feature representing the invariant semantic content, i.e. content feature. Previous methods separate content feature apart by utilizing it to classify haze image during the training process. However, in this paper we recognize the incompleteness of the content-style disentanglement in such technical routine. The flawed style feature entangled with content information inevitably leads the ill-rendering of the haze images. To address, we propose a self-supervised style regression via stochastic linear interpolation to reduce the content information in style feature. The ablative experiments demonstrate the disentangling completeness and its superiority in level-aware haze image synthesis. Moreover, the generated haze data are applied in the testing generalization of vehicle detectors. Further study between haze-level and detection performance shows that haze has obvious impact on the generalization of the vehicle detectors and such performance degrading level is linearly correlated to the haze-level, which, in turn, validates the effectiveness of the proposed method.

2021-11-29

2103.06506

Corey Lammie

Corey Lammie, Jason K. Eshraghian, Wei D. Lu, Mostafa Rahimi Azghadi

Memristive Stochastic Computing for Deep Learning Parameter Optimization

Accepted by IEEE Transactions on Circuits and Systems Part II: Express Briefs

IEEE Transactions on Circuits and Systems Part II: Express Briefs, 2021

10.1109/TCSII.2021.3065932

cs.ET cs.AI cs.AR cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Stochastic Computing (SC) is a computing paradigm that allows for the low-cost and low-power computation of various arithmetic operations using stochastic bit streams and digital logic. In contrast to conventional representation schemes used within the binary domain, the sequence of bit streams in the stochastic domain is inconsequential, and computation is usually non-deterministic. In this brief, we exploit the stochasticity during switching of probabilistic Conductive Bridging RAM (CBRAM) devices to efficiently generate stochastic bit streams in order to perform Deep Learning (DL) parameter optimization, reducing the size of Multiply and Accumulate (MAC) units by 5 orders of magnitude. We demonstrate that in using a 40-nm Complementary Metal Oxide Semiconductor (CMOS) process our scalable architecture occupies 1.55mm$^2$ and consumes approximately 167$\mu$W when optimizing parameters of a Convolutional Neural Network (CNN) while it is being trained for a character recognition task, observing no notable reduction in accuracy post-training.

2021-03-18

2103.06511

Shaoxiong Ji

Shaoxiong Ji and Matti H\"oltt\"a and Pekka Marttinen

Does the Magic of BERT Apply to Medical Code Assignment? A Quantitative Study

Computers in Biology and Medicine, 2021

10.1016/j.compbiomed.2021.104998

cs.CL

http://creativecommons.org/licenses/by/4.0/

Unsupervised pretraining is an integral part of many natural language processing systems, and transfer learning with language models has achieved remarkable results in many downstream tasks. In the clinical application of medical code assignment, diagnosis and procedure codes are inferred from lengthy clinical notes such as hospital discharge summaries. However, it is not clear if pretrained models are useful for medical code prediction without further architecture engineering. This paper conducts a comprehensive quantitative analysis of various contextualized language models' performance, pretrained in different domains, for medical code assignment from clinical notes. We propose a hierarchical fine-tuning architecture to capture interactions between distant words and adopt label-wise attention to exploit label information. Contrary to current trends, we demonstrate that a carefully trained classical CNN outperforms attention-based models on a MIMIC-III subset with frequent codes. Our empirical findings suggest directions for improving the medical code assignment application.

2022-06-03

2103.06552

Theodoros Georgiou

Theodoros Georgiou, Sebastian Schmitt, Thomas B\"ack, Nan Pu, Wei Chen, Michael Lew

PREPRINT: Comparison of deep learning and hand crafted features for mining simulation data

Proceedings of the International Conference on Pattern Recognition (ICPR) 2020

cs.CV

http://creativecommons.org/licenses/by-nc-sa/4.0/

Computational Fluid Dynamics (CFD) simulations are a very important tool for many industrial applications, such as aerodynamic optimization of engineering designs like cars shapes, airplanes parts etc. The output of such simulations, in particular the calculated flow fields, are usually very complex and hard to interpret for realistic three-dimensional real-world applications, especially if time-dependent simulations are investigated. Automated data analysis methods are warranted but a non-trivial obstacle is given by the very large dimensionality of the data. A flow field typically consists of six measurement values for each point of the computational grid in 3D space and time (velocity vector values, turbulent kinetic energy, pressure and viscosity). In this paper we address the task of extracting meaningful results in an automated manner from such high dimensional data sets. We propose deep learning methods which are capable of processing such data and which can be trained to solve relevant tasks on simulation data, i.e. predicting drag and lift forces applied on an airfoil. We also propose an adaptation of the classical hand crafted features known from computer vision to address the same problem and compare a large variety of descriptors and detectors. Finally, we compile a large dataset of 2D simulations of the flow field around airfoils which contains 16000 flow fields with which we tested and compared approaches. Our results show that the deep learning-based methods, as well as hand crafted feature based approaches, are well-capable to accurately describe the content of the CFD simulation output on the proposed dataset.

2021-03-12

2103.06583

Theodoros Georgiou

Theodoros Georgiou, Sebastian Schmitt, Thomas B\"ack, Wei Chen, Michael Lew

Preprint: Norm Loss: An efficient yet effective regularization method for deep neural networks

Proceedings of the International Conference on Pattern Recognition (ICPR) 2020

cs.CV

http://creativecommons.org/licenses/by-nc-sa/4.0/

Convolutional neural network training can suffer from diverse issues like exploding or vanishing gradients, scaling-based weight space symmetry and covariant-shift. In order to address these issues, researchers develop weight regularization methods and activation normalization methods. In this work we propose a weight soft-regularization method based on the Oblique manifold. The proposed method uses a loss function which pushes each weight vector to have a norm close to one, i.e. the weight matrix is smoothly steered toward the so-called Oblique manifold. We evaluate our method on the very popular CIFAR-10, CIFAR-100 and ImageNet 2012 datasets using two state-of-the-art architectures, namely the ResNet and wide-ResNet. Our method introduces negligible computational overhead and the results show that it is competitive to the state-of-the-art and in some cases superior to it. Additionally, the results are less sensitive to hyperparameter settings such as batch size and regularization factor.

2021-03-12

2103.06627

Qiang Meng

Qiang Meng, Shichao Zhao, Zhida Huang, Feng Zhou

MagFace: A Universal Representation for Face Recognition and Quality Assessment

accepted at CVPR 2021, Oral

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The performance of face recognition system degrades when the variability of the acquired faces increases. Prior work alleviates this issue by either monitoring the face quality in pre-processing or predicting the data uncertainty along with the face feature. This paper proposes MagFace, a category of losses that learn a universal feature embedding whose magnitude can measure the quality of the given face. Under the new loss, it can be proven that the magnitude of the feature embedding monotonically increases if the subject is more likely to be recognized. In addition, MagFace introduces an adaptive mechanism to learn a wellstructured within-class feature distributions by pulling easy samples to class centers while pushing hard samples away. This prevents models from overfitting on noisy low-quality samples and improves face recognition in the wild. Extensive experiments conducted on face recognition, quality assessments as well as clustering demonstrate its superiority over state-of-the-arts. The code is available at https://github.com/IrvingMeng/MagFace.

2021-07-27

2103.06752

Daniel Vollmers

Daniel Vollmers (1), Rricha Jalota (1), Diego Moussallem (1), Hardik Topiwala (1), Axel-Cyrille Ngonga Ngomo (1), and Ricardo Usbeck (2) ((1) Data Science Group, Paderborn University, Germany, (2) Fraunhofer IAIS, Dresden, Germany)

Knowledge Graph Question Answering using Graph-Pattern Isomorphism

Version published in the proceedings of the 17th International Conference on Semantic Systems

Further with Knowledge Graphs - Proceedings of the 17th International Conference on Semantic Systems 53 (2021) 103-117

10.3233/SSW210038

cs.AI cs.CL

http://creativecommons.org/licenses/by/4.0/

Knowledge Graph Question Answering (KGQA) systems are based on machine learning algorithms, requiring thousands of question-answer pairs as training examples or natural language processing pipelines that need module fine-tuning. In this paper, we present a novel QA approach, dubbed TeBaQA. Our approach learns to answer questions based on graph isomorphisms from basic graph patterns of SPARQL queries. Learning basic graph patterns is efficient due to the small number of possible patterns. This novel paradigm reduces the amount of training data necessary to achieve state-of-the-art performance. TeBaQA also speeds up the domain adaption process by transforming the QA system development task into a much smaller and easier data compilation task. In our evaluation, TeBaQA achieves state-of-the-art performance on QALD-8 and delivers comparable results on QALD-9 and LC-QuAD v1. Additionally, we performed a fine-grained evaluation on complex queries that deal with aggregation and superlative questions as well as an ablation study, highlighting future research challenges.

2022-02-03

2103.06769

Pierre-Yves Oudeyer

Manfred Eppe and Pierre-Yves Oudeyer

Intelligent behavior depends on the ecological niche: Scaling up AI to human-like intelligence in socio-cultural environments

Keywords: developmental AI, general artificial intelligence, human-like AI, embodiment, cultural evolution, language, socio-cultural skills

KI - K\"unstliche Intelligenz KI - K\"unstliche Intelligenz (German Journal of Artificial Intelligence), 2021

10.1007/s13218-020-00696-1

cs.AI cs.CL cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper outlines a perspective on the future of AI, discussing directions for machines models of human-like intelligence. We explain how developmental and evolutionary theories of human cognition should further inform artificial intelligence. We emphasize the role of ecological niches in sculpting intelligent behavior, and in particular that human intelligence was fundamentally shaped to adapt to a constantly changing socio-cultural environment. We argue that a major limit of current work in AI is that it is missing this perspective, both theoretically and experimentally. Finally, we discuss the promising approach of developmental artificial intelligence, modeling infant development through multi-scale interaction between intrinsically motivated learning, embodiment and a fastly changing socio-cultural environment. This paper takes the form of an interview of Pierre-Yves Oudeyer by Mandred Eppe, organized within the context of a KI - K{\"{u}}nstliche Intelligenz special issue in developmental robotics.

2021-03-12

2103.06854

Laura Giordano

Laura Giordano, Valentina Gliozzi, Daniele Theseider Dupr\'e

A conditional, a fuzzy and a probabilistic interpretation of self-organising maps

31 pages, 1 figure. arXiv admin note: text overlap with arXiv:2008.13278

Journal of Logic and Computation, 2022

10.1093/logcom/exab082

cs.AI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this paper we establish a link between fuzzy and preferential semantics for description logics and Self-Organising Maps, which have been proposed as possible candidates to explain the psychological mechanisms underlying category generalisation. In particular, we show that the input/output behavior of a Self-Organising Map after training can be described by a fuzzy description logic interpretation as well as by a preferential interpretation, based on a concept-wise multipreference semantics, which takes into account preferences with respect to different concepts and has been recently proposed for ranked and for weighted defeasible description logics. Properties of the network can be proven by model checking on the fuzzy or on the preferential interpretation. Starting from the fuzzy interpretation, we also provide a probabilistic account for this neural network model.

2022-02-07

2103.06874

Jonathan Clark

Jonathan H. Clark, Dan Garrette, Iulia Turc, John Wieting

CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation

TACL Final Version

Transactions of the Association for Computational Linguistics (2022) 10: 73--91

10.1162/tacl_a_00448

cs.CL cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Pipelined NLP systems have largely been superseded by end-to-end neural modeling, yet nearly all commonly-used models still require an explicit tokenization step. While recent tokenization approaches based on data-derived subword lexicons are less brittle than manually engineered tokenizers, these techniques are not equally suited to all languages, and the use of any fixed vocabulary may limit a model's ability to adapt. In this paper, we present CANINE, a neural encoder that operates directly on character sequences, without explicit tokenization or vocabulary, and a pre-training strategy that operates either directly on characters or optionally uses subwords as a soft inductive bias. To use its finer-grained input effectively and efficiently, CANINE combines downsampling, which reduces the input sequence length, with a deep transformer stack, which encodes context. CANINE outperforms a comparable mBERT model by 2.8 F1 on TyDi QA, a challenging multilingual benchmark, despite having 28% fewer model parameters.

2022-05-19

2103.06911

Qiaojun Feng

Tianyu Zhao, Qiaojun Feng, Sai Jadhav, Nikolay Atanasov

CORSAIR: Convolutional Object Retrieval and Symmetry-AIded Registration

8 pages, 8 figures

2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Prague, Czech Republic, 2021, pp. 47-54

10.1109/IROS51168.2021.9636347

cs.CV cs.RO

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper considers online object-level mapping using partial point-cloud observations obtained online in an unknown environment. We develop and approach for fully Convolutional Object Retrieval and Symmetry-AIded Registration (CORSAIR). Our model extends the Fully Convolutional Geometric Features model to learn a global object-shape embedding in addition to local point-wise features from the point-cloud observations. The global feature is used to retrieve a similar object from a category database, and the local features are used for robust pose registration between the observed and the retrieved object. Our formulation also leverages symmetries, present in the object shapes, to obtain promising local-feature pairs from different symmetry classes for matching. We present results from synthetic and real-world datasets with different object categories to verify the robustness of our method.

2022-04-26

2103.07156

Kohei Yamamoto

Kohei Yamamoto

Learnable Companding Quantization for Accurate Low-bit Neural Networks

Accepted at CVPR 2021

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 5029-5038

cs.CV cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Quantizing deep neural networks is an effective method for reducing memory consumption and improving inference speed, and is thus useful for implementation in resource-constrained devices. However, it is still hard for extremely low-bit models to achieve accuracy comparable with that of full-precision models. To address this issue, we propose learnable companding quantization (LCQ) as a novel non-uniform quantization method for 2-, 3-, and 4-bit models. LCQ jointly optimizes model weights and learnable companding functions that can flexibly and non-uniformly control the quantization levels of weights and activations. We also present a new weight normalization technique that allows more stable training for quantization. Experimental results show that LCQ outperforms conventional state-of-the-art methods and narrows the gap between quantized and full-precision models for image classification and object detection tasks. Notably, the 2-bit ResNet-50 model on ImageNet achieves top-1 accuracy of 75.1% and reduces the gap to 1.7%, allowing LCQ to further exploit the potential of non-uniform quantization.

2021-11-03

2103.07202

Cl\'ement Rambour

Cl\'ement Rambour, Lo\"ic Denis, Florence Tupin, H\'el\`ene Oriot, Yue Huang, Laurent Ferro-Famil

Urban Surface Reconstruction in SAR Tomography by Graph-Cuts

Computer Vision and Image Understanding 188 (2019) 102791

10.1016/j.cviu.2019.07.011

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

SAR (Synthetic Aperture Radar) tomography reconstructs 3-D volumes from stacks of SAR images. High-resolution satellites such as TerraSAR-X provide images that can be combined to produce 3-D models. In urban areas, sparsity priors are generally enforced during the tomographic inversion process in order to retrieve the location of scatterers seen within a given radar resolution cell. However, such priors often miss parts of the urban surfaces. Those missing parts are typically regions of flat areas such as ground or rooftops. This paper introduces a surface segmentation algorithm based on the computation of the optimal cut in a flow network. This segmentation process can be included within the 3-D reconstruction framework in order to improve the recovery of urban surfaces. Illustrations on a TerraSAR-X tomographic dataset demonstrate the potential of the approach to produce a 3-D model of urban surfaces such as ground, fa\c{c}ades and rooftops.

2021-03-15

2103.07278

Julien Despois

Hugo Thimonier, Julien Despois, Robin Kips, Matthieu Perrot

Learning Long-Term Style-Preserving Blind Video Temporal Consistency

2021 IEEE International Conference on Multimedia and Expo (ICME)

10.1109/ICME51207.2021.9428445

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

When trying to independently apply image-trained algorithms to successive frames in videos, noxious flickering tends to appear. State-of-the-art post-processing techniques that aim at fostering temporal consistency, generate other temporal artifacts and visually alter the style of videos. We propose a postprocessing model, agnostic to the transformation applied to videos (e.g. style transfer, image manipulation using GANs, etc.), in the form of a recurrent neural network. Our model is trained using a Ping Pong procedure and its corresponding loss, recently introduced for GAN video generation, as well as a novel style preserving perceptual loss. The former improves long-term temporal consistency learning, while the latter fosters style preservation. We evaluate our model on the DAVIS and videvo.net datasets and show that our approach offers state-of-the-art results concerning flicker removal, and better keeps the overall style of the videos than previous approaches.

2022-10-06

2103.07492

Andrea Cossu

Andrea Cossu, Antonio Carta, Vincenzo Lomonaco, Davide Bacciu

Continual Learning for Recurrent Neural Networks: an Empirical Evaluation

Published in Neural Networks

Neural Networks, Volume 143, 2021, pages 607-627

10.1016/j.neunet.2021.07.021

cs.LG cs.AI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Learning continuously during all model lifetime is fundamental to deploy machine learning solutions robust to drifts in the data distribution. Advances in Continual Learning (CL) with recurrent neural networks could pave the way to a large number of applications where incoming data is non stationary, like natural language processing and robotics. However, the existing body of work on the topic is still fragmented, with approaches which are application-specific and whose assessment is based on heterogeneous learning protocols and datasets. In this paper, we organize the literature on CL for sequential data processing by providing a categorization of the contributions and a review of the benchmarks. We propose two new benchmarks for CL with sequential data based on existing datasets, whose characteristics resemble real-world applications. We also provide a broad empirical evaluation of CL and Recurrent Neural Networks in class-incremental scenario, by testing their ability to mitigate forgetting with a number of different strategies which are not specific to sequential data processing. Our results highlight the key role played by the sequence length and the importance of a clear specification of the CL scenario.

2021-08-03

2103.07538

Sandeep Soni

Sandeep Soni and Lauren Klein and Jacob Eisenstein

Abolitionist Networks: Modeling Language Change in Nineteenth-Century Activist Newspapers

23 pages, 6 figures, 2 tables

Journal of Cultural Analytics (2021)

cs.CL cs.CY cs.DL cs.SI

http://creativecommons.org/licenses/by/4.0/

The abolitionist movement of the nineteenth-century United States remains among the most significant social and political movements in US history. Abolitionist newspapers played a crucial role in spreading information and shaping public opinion around a range of issues relating to the abolition of slavery. These newspapers also serve as a primary source of information about the movement for scholars today, resulting in powerful new accounts of the movement and its leaders. This paper supplements recent qualitative work on the role of women in abolition's vanguard, as well as the role of the Black press, with a quantitative text modeling approach. Using diachronic word embeddings, we identify which newspapers tended to lead lexical semantic innovations -- the introduction of new usages of specific words -- and which newspapers tended to follow. We then aggregate the evidence across hundreds of changes into a weighted network with the newspapers as nodes; directed edge weights represent the frequency with which each newspaper led the other in the adoption of a lexical semantic change. Analysis of this network reveals pathways of lexical semantic influence, distinguishing leaders from followers, as well as others who stood apart from the semantic changes that swept through this period. More specifically, we find that two newspapers edited by women -- THE PROVINCIAL FREEMAN and THE LILY -- led a large number of semantic changes in our corpus, lending additional credence to the argument that a multiracial coalition of women led the abolitionist movement in terms of both thought and action. It also contributes additional complexity to the scholarship that has sought to tease apart the relation of the abolitionist movement to the women's suffrage movement, and the vexed racial politics that characterized their relation.

2021-03-16

2103.07609

Kristina Monakhova

Kristina Monakhova, Vi Tran, Grace Kuo, Laura Waller

Untrained networks for compressive lensless photography

17 pages, 8 figures

Optics Express Vol. 29, Issue 13, pp. 20913-20929 (2021)

10.1364/OE.424075

eess.IV cs.CV physics.optics

http://creativecommons.org/licenses/by/4.0/

Compressive lensless imagers enable novel applications in an extremely compact device, requiring only a phase or amplitude mask placed close to the sensor. They have been demonstrated for 2D and 3D microscopy, single-shot video, and single-shot hyperspectral imaging; in each of these cases, a compressive-sensing-based inverse problem is solved in order to recover a 3D data-cube from a 2D measurement. Typically, this is accomplished using convex optimization and hand-picked priors. Alternatively, deep learning-based reconstruction methods offer the promise of better priors, but require many thousands of ground truth training pairs, which can be difficult or impossible to acquire. In this work, we propose the use of untrained networks for compressive image recovery. Our approach does not require any labeled training data, but instead uses the measurement itself to update the network weights. We demonstrate our untrained approach on lensless compressive 2D imaging as well as single-shot high-speed video recovery using the camera's rolling shutter, and single-shot hyperspectral imaging. We provide simulation and experimental verification, showing that our method results in improved image quality over existing methods.

2021-06-23

2103.07612

Matloob Khushi Dr

Mimi Mukherjee and Matloob Khushi

SMOTE-ENC: A novel SMOTE-based method to generate synthetic data for nominal and continuous features

Appl. Syst. Innov. 2021, 4, 18

10.3390/asi4010018

cs.LG cs.AI

http://creativecommons.org/licenses/by/4.0/

Real world datasets are heavily skewed where some classes are significantly outnumbered by the other classes. In these situations, machine learning algorithms fail to achieve substantial efficacy while predicting these under-represented instances. To solve this problem, many variations of synthetic minority over-sampling methods (SMOTE) have been proposed to balance the dataset which deals with continuous features. However, for datasets with both nominal and continuous features, SMOTE-NC is the only SMOTE-based over-sampling technique to balance the data. In this paper, we present a novel minority over-sampling method, SMOTE-ENC (SMOTE - Encoded Nominal and Continuous), in which, nominal features are encoded as numeric values and the difference between two such numeric value reflects the amount of change of association with minority class. Our experiments show that the classification model using SMOTE-ENC method offers better prediction than model using SMOTE-NC when the dataset has a substantial number of nominal features and also when there is some association between the categorical features and the target class. Additionally, our proposed method addressed one of the major limitations of SMOTE-NC algorithm. SMOTE-NC can be applied only on mixed datasets that have features consisting of both continuous and nominal features and cannot function if all the features of the dataset are nominal. Our novel method has been generalized to be applied on both mixed datasets and on nominal only datasets. The code is available from mkhushi.github.io

2021-03-16

2103.07678

Ehsan Farahbakhsh

Hojat Shirmard, Ehsan Farahbakhsh, R. Dietmar Muller, Rohitash Chandra

A review of machine learning in processing remote sensing data for mineral exploration

26 pages, 4 figures, 2 tables

Remote Sensing of Environment, 268, 112750 (2022)

10.1016/j.rse.2021.112750

cs.LG cs.CV stat.AP

http://creativecommons.org/licenses/by/4.0/

The decline of the number of newly discovered mineral deposits and increase in demand for different minerals in recent years has led exploration geologists to look for more efficient and innovative methods for processing different data types at each stage of mineral exploration. As a primary step, various features, such as lithological units, alteration types, structures, and indicator minerals, are mapped to aid decision-making in targeting ore deposits. Different types of remote sensing datasets, such as satellite and airborne data, make it possible to overcome common problems associated with mapping geological features. The rapid increase in the volume of remote sensing data obtained from different platforms has encouraged scientists to develop advanced, innovative, and robust data processing methodologies. Machine learning methods can help process a wide range of remote sensing datasets and determine the relationship between components such as the reflectance continuum and features of interest. These methods are robust in processing spectral and ground truth measurements against noise and uncertainties. In recent years, many studies have been carried out by supplementing geological surveys with remote sensing datasets, which is now prominent in geoscience research. This paper provides a comprehensive review of the implementation and adaptation of some popular and recently established machine learning methods for processing different types of remote sensing data and investigates their applications for detecting various ore deposit types. We demonstrate the high capability of combining remote sensing data and machine learning methods for mapping different geological features that are critical for providing potential maps. Moreover, we find there is scope for advanced methods to process the new generation of remote sensing data for creating improved mineral prospectivity maps.

2021-12-07

2103.07762

Bonaventure F. P. Dossou

Bonaventure F. P. Dossou and Chris C. Emezue

OkwuGb\'e: End-to-End Speech Recognition for Fon and Igbo

African NLP, EACL 2021

cs.CL cs.AI cs.CY

http://creativecommons.org/licenses/by/4.0/

Language is inherent and compulsory for human communication. Whether expressed in a written or spoken way, it ensures understanding between people of the same and different regions. With the growing awareness and effort to include more low-resourced languages in NLP research, African languages have recently been a major subject of research in machine translation, and other text-based areas of NLP. However, there is still very little comparable research in speech recognition for African languages. Interestingly, some of the unique properties of African languages affecting NLP, like their diacritical and tonal complexities, have a major root in their speech, suggesting that careful speech interpretation could provide more intuition on how to deal with the linguistic complexities of African languages for text-based NLP. OkwuGb\'e is a step towards building speech recognition systems for African low-resourced languages. Using Fon and Igbo as our case study, we conduct a comprehensive linguistic analysis of each language and describe the creation of end-to-end, deep neural network-based speech recognition models for both languages. We present a state-of-art ASR model for Fon, as well as benchmark ASR model results for Igbo. Our linguistic analyses (for Fon and Igbo) provide valuable insights and guidance into the creation of speech recognition models for other African low-resourced languages, as well as guide future NLP research for Fon and Igbo. The Fon and Igbo models source code have been made publicly available.

2021-03-17

2103.07768

Robin Swezey

Robin Swezey, Bruno Charron

Large-scale Recommendation for Portfolio Optimization

In Proceedings of the 12th ACM Conference on Recommender Systems (RecSys 2018). Association for Computing Machinery, New York, NY, USA, 382-386

10.1145/3240323.3240386

cs.AI cs.CE cs.IR cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Individual investors are now massively using online brokers to trade stocks with convenient interfaces and low fees, albeit losing the advice and personalization traditionally provided by full-service brokers. We frame the problem faced by online brokers of replicating this level of service in a low-cost and automated manner for a very large number of users. Because of the care required in recommending financial products, we focus on a risk-management approach tailored to each user's portfolio and risk profile. We show that our hybrid approach, based on Modern Portfolio Theory and Collaborative Filtering, provides a sound and effective solution. The method is applicable to stocks as well as other financial assets, and can be easily combined with various financial forecasting models. We validate our proposal by comparing it with several baselines in a domain expert-based study.

2021-03-16