Datasets:

debajyotidasgupta

/

dblp_2022_2023

like 0

Title: An Edge Computing-Based Photo Crowdsourcing Framework for Real-Time 3D Reconstruction Abstract: Image-based three-dimensional (3D) reconstruction utilizes a set of photos to build 3D model and can be widely used in many emerging applications such as augmented reality (AR) and disaster recovery. Most of existing 3D reconstruction methods require a mobile user to walk around the target area and reconstruct objectives with a hand-held camera, which is inefficient and time-consuming. To meet the...

Title: Stabilization of a class of underactuated Euler Lagrange system using an approximate model Abstract: The energy shaping method, Controlled Lagrangian, is a well-known approach to stabilize the underactuated Euler Lagrange (EL) systems. In this approach, to construct a control rule, some nonlinear and nonhomogeneous partial differential equations (PDEs), which are called matching conditions, must be solved. In this paper, a method is proposed to obtain an approximate solution of these matching conditions for a class of underactuated EL systems. To develop this method, the potential energy matching condition is transformed to a set of linear PDEs using an approximation of inertia matrices. Hence, the assignable potential energy function and the controlled inertia matrix both are constructed as a common solution of these PDEs. Subsequently, the gyroscopic and dissipative forces are determined as the solution for kinetic energy matching condition. Conclusively, the control rule is constructed by adding energy shaping rule and additional dissipation injection to provide asymptotic stability. The stability analysis of the closed-loop system which used the control rule derived with the proposed method is also provided. To demonstrate the success of the proposed method, the stability problem of the inverted pendulum on a cart is considered.

Title: Choiceless large cardinals and set-theoretic potentialism Abstract: We define a potentialist system of ZF$\mathsf {ZF}$-structures, i.e., a collection of possible worlds in the language of ZF$\mathsf {ZF}$ connected by a binary accessibility relation, achieving a potentialist account of the full background set-theoretic universe V. The definition involves Berkeley cardinals, the strongest known large cardinal axioms, inconsistent with the Axiom of Choice. In fact, as background theory we assume just ZF$\mathsf {ZF}$. It turns out that the propositional modal assertions which are valid at every world of our system are exactly those in the modal theory S4.2$\mathsf {S4.2}$. Moreover, we characterize the worlds satisfying the potentialist maximality principle, and thus the modal theory S5$\mathsf {S5}$, both for assertions in the language of ZF$\mathsf {ZF}$ and for assertions in the full potentialist language.

Title: A Machine Learning Pipeline Stage for Adaptive Frequency Adjustment Abstract: A machine learning (ML) design framework is proposed for adaptively adjusting clock frequency based on propagation delay of individual instructions. A random forest model is trained to classify propagation delays in real time, utilizing current operation type, current operands, and computation history as ML features. The trained model is implemented in Verilog as an additional pipeline stage within TigerMIPS processor. The modified system is experimentally tested at the gate level in 45 nm CMOS technology, exhibiting simultaneously a speedup of 70 percent and an energy reduction of 30 percent with coarse-grained ML classification as compared with the baseline TigerMIPS. A speedup of 89 percent is demonstrated with finer granularities with a simultaneous 15.5 percent reduction in energy consumption.

Title: Language-agnostic BERT Sentence Embedding Abstract: While BERT is an effective method for learning monolingual sentence embeddings for semantic similarity and embedding based transfer learning (Reimers and Gurevych, 2019), BERT based cross-lingual sentence embeddings have yet to be explored. We systematically investigate methods for learning multilingual sentence embeddings by combining the best methods for learning monolingual and cross-lingual representations including: masked language modeling (MLM), translation language modeling (TLM) (Conneau and Lample, 2019), dual encoder translation ranking (Guo et al., 2018), and additive margin softmax (Yang et al., 2019a). We show that introducing a pre-trained multilingual language model dramatically reduces the amount of parallel training data required to achieve good performance by 80%. Composing the best of these methods produces a model that achieves 83.7% bi-text retrieval accuracy over 112 languages on Tatoeba, well above the 65.5% achieved by Artetxe and Schwenk (2019b), while still performing competitively on monolingual transfer learning benchmarks (Conneau and Kiela, 2018). Parallel data mined from CommonCrawl using our best model is shown to train competitive NMT models for en-zh and en-de. We publicly release our best multilingual sentence embedding model for 109+ languages at https://tfhub.dev/google/LaBSE.

Title: Perfect matchings in highly cyclically connected regular graphs Abstract: A leaf matching operation on a graph consists of removing a vertex of degree 1 together with its neighbour from the graph. Let G be a d-regular cyclically ( d - 1 + 2 k )-edge-connected graph of even order, where k >= 0 and d >= 3. We prove that for any given set X of d - 1 + k edges, there is no 1-factor of G avoiding X if and only if either an isolated vertex can be obtained by a series of leaf matching operations in G - X, or G - X has an independent set that contains more than half of the vertices of G. To demonstrate how to check the conditions of the theorem we prove several statements on 2-factors of cubic graphs. For k >= 3, we prove that given a cyclically ( 4 k - 5 )-edge-connected cubic graph G and three paths of length k such that the distance between any two of them is at least 8 k - 16, there is a 2-factor of G that contains one of the paths. We provide a similar statement for two paths when k = 3 and k = 4. As a corollary we show that given a vertex v in a cyclically 7-edge-connected cubic graph, there is a 2-factor such that v is in a circuit of length greater than 7.

Title: Classes of graphs without star forests and related graphs Abstract: This work provides a structural characterisation of hereditary graph classes that do not contain a star forest, several graphs obtained from star forests by subset complementation, a union of cliques, and the complement of a union of cliques as induced subgraphs. This provides, for instance, structural results for graph classes not containing a matching and several complements of a matching. In terms of the speed of hereditary graph classes, our results imply that all such classes have at most factorial speed of growth.

Title: Cross-Network Skip-Gram Embedding for Joint Network Alignment and Link Prediction Abstract: Link prediction and network alignment are two fundamental and interleaved tasks in network analysis. In this paper, we propose a novel cross-network embedding model under the Skip-gram framework, which alternately performs link prediction and network alignment by joint optimization. Vertex sequences, obtained via a biased random walk based on empirical mixture distributions, are used to train a Skip-gram based node embedding model. On one hand, based on the similarity in embedding space, network alignment can be effectively performed either with the initial ground truth alignments as seeds or from scratch. On the other hand, the proposed link prediction model involves training a supervised classifier by sampling a set of positive and negative edges. We also modify and incorporate the Collective Link Fusion (CLF) method under a Skip-gram framework and show that the new method can achieve better results in both tasks. Extensive experimental results show the state-of-the-art performance of our methods.

Title: $\mathsf{NCF}$ NCF : A Neural Context Fusion Approach to Raw Mobility Annotation Abstract: Understanding human mobility patterns at the point-of-interest (POI) scale plays an important role in enhancing business intelligence in mobile environments. While large efforts have been made in this direction, most studies simply utilize POI check-ins to mine the concerned mobility patterns, the effectiveness of which is usually hindered due to data sparsity. To obtain better POI-based human mob...

Title: STMARL: A Spatio-Temporal Multi-Agent Reinforcement Learning Approach for Cooperative Traffic Light Control Abstract: The development of intelligent traffic light control systems is essential for smart transportation management. While some efforts have been made to optimize the use of individual traffic lights in an isolated way, related studies have largely ignored the fact that the use of multi-intersection traffic lights is spatially influenced, as well as the temporal dependency of historical traffic status f...

Title: Domain-Adversarial Network Alignment Abstract: Network alignment is a critical task in a wide variety of fields. Many existing works leverage on representation learning to accomplish this task without eliminating domain representation bias induced by domain-dependent features, which yield inferior alignment performance. This paper proposes a unified deep architecture ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">DANA</i> ) to obtain a domain-invariant representation for network alignment via an adversarial domain classifier. Specifically, we employ the graph convolutional networks to perform network embedding under the domain adversarial principle, given a small set of observed anchors. Then, the semi-supervised learning framework is optimized by maximizing a posterior probability distribution of observed anchors and the loss of a domain classifier simultaneously. We also develop a few variants of our model, such as, direction-aware network alignment, weight-sharing for directed networks and simplification of parameter space. Experiments on three real-world social network datasets demonstrate that our proposed approaches achieve state-of-the-art alignment results.

Title: Dynamic Type Matching Abstract: Problem definition: We consider an intermediary's problem of dynamically matching demand and supply of heterogeneous types in a periodic-review fashion. Specifically, there are two disjoint sets of demand and supply types, and a reward for each possible matching of a demand type and a supply type. In each period, demand and supply of various types arrive in random quantities. The platform decides on the optimal matching policy to maximize the expected total discounted rewards, given that unmatched demand and supply may incur waiting or holding costs, and will be fully or partially carried over to the next period. Academic/practical relevance: The problem is crucial to many intermediaries who manage matchings centrally in a sharing economy. Methodology: We formulate the problem as a dynamic program. We explore the structural properties of the optimal policy and propose heuristic policies. Results: We provide sufficient conditions on matching rewards such that the optimal matching policy follows a priority hierarchy among possible matching pairs. We show that those conditions are satisfied by vertically and unidirectionally horizontally differentiated types, for which quality and distance determine priority, respectively. Managerial implications: The priority property simplifies the matching decision within a period, and the trade-off reduces to a choice between matching in the current period and that in the future. Then the optimal matching policy has a match-down-to structure when considering a specific pair of demand and supply types in the priority hierarchy.

Title: Enhanced Discrete Multi-Modal Hashing: More Constraints Yet Less Time to Learn Abstract: Due to the exponential growth of multimedia data, multi-modal hashing as a promising technique to make cross-view retrieval scalable is attracting more and more attention. However, most of the existing multi-modal hashing methods either divide the learning process unnaturally into two separate stages or treat the discrete optimization problem simplistically as a continuous one, which leads to suboptimal results. Recently, a few discrete multi-modal hashing methods that try to address such issues have emerged, but they still ignore several important discrete constraints (such as the balance and decorrelation of hash bits). In this paper, we overcome those limitations by proposing a novel method named “Enhanced Discrete Multi-modal Hashing (EDMH)” which learns binary codes and hashing functions simultaneously from the pairwise similarity matrix of data, under the aforementioned discrete constraints. Although the model of EDMH looks a lot more complex than the other models for multi-modal hashing, we are actually able to develop a fast iterative learning algorithm for it, since the subproblems of its optimization all have closed-form solutions after introducing a couple of auxiliary variables. Our experimental results on three real-world datasets have revealed the usefulness of those previously ignored discrete constraints and demonstrated that EDMH not only performs much better than state-of-the-art competitors according to several retrieval metrics but also runs much faster than most of them.

Title: Graphical Convergence of Subgradien s in Nonconvex Optimization and Learning Abstract: We investigate the stochastic optimization problem of minimizing population risk, where the loss defining the risk is assumed to be weakly convex. Compositions of Lipschitz convex functions with smooth maps are the primary examples of such losses. We analyze the estimation quality of such nonsmooth and nonconvex problems by their sample average approximations. Our main results establish dimension-dependent rates on subgradient estimation in full generality and dimension-independent rates when the loss is a generalized linear model. As an application of the developed techniques, we analyze the nonsmooth landscape of a robust nonlinear regression problem.

Title: Equivalence of generics Abstract: Given a countable transitive model of set theory and a partial order contained in it, there is a natural countable Borel equivalence relation on generic filters over the model; two are equivalent if they yield the same generic extension. We examine the complexity of this equivalence relation for various partial orders, focusing on Cohen and random forcing. We prove, among other results, that the former is an increasing union of countably many hyperfinite Borel equivalence relations, and hence is amenable, while the latter is neither amenable nor treeable.

Title: BiFNet: Bidirectional Fusion Network for Road Segmentation Abstract: Multisensor fusion-based road segmentation plays an important role in the intelligent driving system since it provides a drivable area. The existing mainstream fusion method is mainly to feature fusion in the image space domain which causes the perspective compression of the road and damages the performance of the distant road. Considering the bird’s eye views (BEVs) of the LiDAR remains the space structure in the horizontal plane, this article proposes a bidirectional fusion network (BiFNet) to fuse the image and BEV of the point cloud. The network consists of two modules: 1) the dense space transformation (DST) module, which solves the mutual conversion between the camera image space and BEV space and 2) the context-based feature fusion module, which fuses the different sensors information based on the scenes from corresponding features. This method has achieved competitive results on the KITTI dataset.

Title: Accurate bounding-box regression with distance-IoU loss for visual tracking Abstract: Most existing trackers are based on using a classifier and multi-scale estimation to estimate the target state. Consequently, and as expected, trackers have become more stable while tracking accuracy has stagnated. While trackers adopt a maximum overlap method based on an intersection-over-union (IoU) loss to mitigate this problem, there are defects in the IoU loss itself, that make it impossible to continue to optimize the objective function when a given bounding box is completely contained within/without another bounding box; this makes it very challenging to accurately estimate the target state. Accordingly, in this paper, we address the above-mentioned problem by proposing a novel tracking method based on a distance-IoU (DIoU) loss, such that the proposed tracker consists of target estimation and target classification. The target estimation part is trained to predict the DIoU score between the target ground-truth bounding-box and the estimated bounding-box. The DIoU loss can maintain the advantage provided by the IoU loss while minimizing the distance between the center points of two bounding boxes, thereby making the target estimation more accurate. Moreover, we introduce a classification part that is trained online and optimized with a Conjugate-Gradient-based strategy to guarantee real-time tracking speed. Comprehensive experimental results demonstrate that the proposed method achieves competitive tracking accuracy when compared to state-of-the-art trackers while with a real-time tracking speed.

Title: New lower bounds for the first variable Zagreb index Abstract: In this paper we study lower bounds in a unified way for a large family of topological indices, including the first variable Zagreb index M-1(alpha). Our aim is to obtain sharp inequalities and characterize the corresponding extremal graphs. The main results provide lower bounds for several vertex-degree-based topological indices. These bounds are new even for the first Zagreb, the inverse and the forgotten indices. (C) 2021 Elsevier B.V. All rights reserved.

Title: A gentle introduction to the differential equation method and dynamic concentration Abstract: We discuss the differential equation method for establishing dynamic concentration of discrete random processes. We present several relatively simple examples of it and aim to make the method understandable to the unfamiliar reader who has some basic knowledge on probabilistic methods, random graphs and differential equations.

Title: Weight-Dependent Gates for Network Pruning Abstract: In this paper, a simple yet effective network pruning framework is proposed to simultaneously address the problems of pruning indicator, pruning ratio, and efficiency constraint. This paper argues that the pruning decision should depend on the convolutional weights, and thus proposes novel weight-dependent gates (W-Gates) to learn the information from filter weights and obtain binary gates to prune or keep the filters automatically. To prune the network under efficiency constraints, a switchable Efficiency Module is constructed to predict the hardware latency or FLOPs of candidate pruned networks. Combined with the proposed Efficiency Module, W-Gates can perform filter pruning in an efficiency-aware manner and achieve a compact network with a better accuracy-efficiency trade-off. We have demonstrated the effectiveness of the proposed method on ResNet34, ResNet50, and MobileNet V2, respectively achieving up to 1.33/1.28/1.1 higher Top-1 accuracy with lower hardware latency on ImageNet. Compared with state-of-the-art methods, W-Gates also achieves superior performance.

Title: Speckle2Void: Deep Self-Supervised SAR Despeckling With Blind-Spot Convolutional Neural Networks Abstract: Information extraction from synthetic aperture radar (SAR) images is heavily impaired by speckle noise, and hence, despeckling is a crucial preliminary step in scene analysis algorithms. The recent success of deep learning envisions a new generation of despeckling techniques that could outperform classical model-based methods. However, current deep learning approaches to despeckling require supervision for training, whereas clean SAR images are impossible to obtain. In the literature, this issue is tackled by resorting to either synthetically speckled optical images, which exhibit different properties with respect to true SAR images, or multitemporal SAR images, which are difficult to acquire or fuse accurately. In this article, inspired by recent works on blind-spot denoising networks, we propose a self-supervised Bayesian despeckling method. The proposed method is trained by employing only noisy SAR images and can, therefore, learn features of real SAR images rather than synthetic data. Experiments show that the performance of the proposed approach is very close to the supervised training approach on synthetic data and superior on real data in both quantitative and visual assessments.

Title: Coded Distributed Computing With Partial Recovery Abstract: Coded computation techniques provide robustness against straggling workers in distributed computing. However, most of the existing schemes require exact provisioning of the straggling behavior and ignore the computations carried out by straggling workers. Moreover, these schemes are typically designed to recover the desired computation results accurately, while in many machine lea...

Title: A Survey on Applications of Artificial Intelligence in Fighting Against COVID-19 Abstract: AbstractThe COVID-19 pandemic caused by the SARS-CoV-2 virus has spread rapidly worldwide, leading to a global outbreak. Most governments, enterprises, and scientific research institutions are participating in the COVID-19 struggle to curb the spread of the pandemic. As a powerful tool against COVID-19, artificial intelligence (AI) technologies are widely used in combating this pandemic. In this survey, we investigate the main scope and contributions of AI in combating COVID-19 from the aspects of disease detection and diagnosis, virology and pathogenesis, drug and vaccine development, and epidemic and transmission prediction. In addition, we summarize the available data and resources that can be used for AI-based COVID-19 research. Finally, the main challenges and potential directions of AI in fighting against COVID-19 are discussed. Currently, AI mainly focuses on medical image inspection, genomics, drug development, and transmission prediction, and thus AI still has great potential in this field. This survey presents medical and AI researchers with a comprehensive view of the existing and potential applications of AI technology in combating COVID-19 with the goal of inspiring researchers to continue to maximize the advantages of AI and big data to fight COVID-19.

Title: Blind inverse gamma correction with maximized differential entropy Abstract: •The blind inverse gamma correction is modeled with maximum entropy assumption.•A closed-form solution via differential entropy and change-of-variables rule.•GCME is an exact (non-approximate), accurate, and fast gamma correction algorithm.•Verified on setting imaging gamma, contrast enhancement, and FPP gamma correction.

Title: Are Chain-Complete Posets Co-wellpowered? Abstract: We show that the category CPO of chain-complete posets is not co-wellpowered but that it is weakly co-wellpowered. This implies that CPO is nearly locally presentable.

Title: Temporal Logic Trees for Model Checking and Control Synthesis of Uncertain Discrete-Time Systems Abstract: We propose algorithms for performing model checking and control synthesis for discrete-time uncertain systems under linear temporal logic (LTL) specifications. We construct temporal logic trees (TLTs) from LTL formulae via reachability analysis. In contrast to automaton-based methods, the construction of the TLT is abstraction-free for infinite systems; that is, we do not construct discrete abstractions of the infinite systems. Moreover, for a given transition system and an LTL formula, we prove that there exist both a universal TLT and an existential TLT via minimal and maximal reachability analysis, respectively. We show that the universal TLT is an underapproximation for the LTL formula and the existential TLT is an overapproximation. We provide sufficient conditions and necessary conditions to verify whether a transition system satisfies an LTL formula by using the TLT approximations. As a major contribution of this work, for a controlled transition system and an LTL formula, we prove that a controlled TLT can be constructed from the LTL formula via a control-dependent reachability analysis. Based on the controlled TLT, we design an online control synthesis algorithm, under which a set of feasible control inputs can be generated at each time step. We also prove that this algorithm is recursively feasible. We illustrate the proposed methods for both finite and infinite systems and highlight the generality and online scalability with two simulated examples.

Title: Complexity of the multilevel critical node problem Abstract: In this work, we analyze a sequential game played in a graph called the Multilevel Critical Node problem (MCN). A defender and an attacker are the players of this game. The defender starts by preventively interdicting vertices (vaccination) from being attacked. Then, the attacker infects a subset of non-vaccinated vertices and, finally, the defender reacts with a protection strategy. We provide the first computational complexity results associated with MCN and its subgames. Moreover, by considering unitary, weighted, undirected, and directed graphs, we clarify how the theoretical tractability of those problems vary. Our findings contribute with new NP-complete, Σ2p-complete and Σ3p-complete problems. Furthermore, for the last level of the game, the protection stage, we build polynomial time algorithms for certain graph classes.

Title: Auto-captions on GIF: A Large-scale Video-sentence Dataset for Vision-language Pre-training Abstract: ABSTRACTIn this work, we present Auto-captions on GIF (ACTION), which is a new large-scale pre-training dataset for generic video understanding. All video-sentence pairs are created by automatically extracting and filtering video caption annotations from billions of web pages. Auto-captions on GIF dataset can be utilized to pre-train the generic feature representation or encoder-decoder structure for video captioning, and other downstream tasks (e.g., sentence localization in videos, video question answering, etc.) as well. We present a detailed analysis of Auto-captions on GIF dataset in comparison to existing video-sentence datasets. We also provide an evaluation of a Transformer-based encoder-decoder structure for vision-language pre-training, which is further adapted to video captioning downstream task and yields the compelling generalizability on MSR-VTT. The dataset is available at http://www.auto-video-captions.top/2022/dataset.

Title: Novel Min-Max Reformulations of Linear Inverse Problems Abstract: In this article, we dwell into the class of so-called ill-posed Linear Inverse Problems (LIP) which simply refer to the task of recovering the entire signal from its relatively few random linear measurements. Such problems arise in a variety of settings with applications ranging from medical image processing, recommender systems, etc. We propose a slightly generalized version of the error constrained linear inverse problem and obtain a novel and equivalent convex-concave min-max reformulation by providing an exposition to its convex geometry. Saddle points of the min-max problem are completely characterized in terms of a solution to the LIP, and vice versa. Applying simple saddle point seeking ascend-descent type algorithms to solve the min-max problems provides novel and simple algorithms to find a solution to the LIP. Moreover, the reformulation of an LIP as the min-max problem provided in this article is crucial in developing methods to solve the dictionary learning problem with almost sure recovery constraints.

Title: Graphop Mean-Field Limits for Kuramoto-Type Models Abstract: Originally arising in the context of interacting particle systems in statistical physics, dynamical systems and differential equations on networks/graphs have permeated into a broad number of mathematical areas as well as into many applications. One central problem in the field is to find suitable approximations of the dynamics as the number of nodes/vertices tends to infinity, i.e., in the large graph limit. A cornerstone in this context are Vlasov equations (VEs) describing a particle density on a mean-field level. For all-to-all coupled systems, it is quite classical to prove the rigorous approximation by VEs for many classes of particle systems. For dense graphs converging to graphon limits, one also knows that mean-field approximation holds for certain classes of models, e.g., for the Kuramoto model on graphs. Yet, the space of intermediate density and sparse graphs is clearly extremely relevant. Here we prove that the Kuramoto model can be be approximated in the mean field limit by far more general graph limits than graphons. In particular, our contributions are as follows. (I) We show how to introduce operator theory more abstractly into VEs by considering graphops. Graphops have recently been proposed as a unifying approach to graph limit theory, and here we show that they can be used for differential equations on graphs. (II) For the Kuramoto model on graphs we rigorously prove that there is a VE approximating it in the mean-field sense. (III) This mean-field VE involves a graphop, and we prove the existence, uniqueness, and continuous graphop dependence of weak solutions. (IV) On a technical level, our results rely on designing a new suitable metric of graphop convergence and on employing Fourier analysis on compact abelian groups to approximate graphops using summability kernels.

Title: Multilinear Weighted Estimates and Quantum Zakharov System Abstract: We consider the well-posedness theory of the compact case of one-dimensional quantum Zakharov system with the periodic boundary condition. The global well-posedness for sufficiently regular data is shown. The semi-classical limit as epsilon -> 0 is obtained on a compact time interval whereas the quantum perturbation proves to be singular on an infinite time interval.

Title: Industrial Forecasting with Exponentially Smoothed Recurrent Neural Networks Abstract: Time series modeling has entered an era of unprecedented growth in the size and complexity of data which require new modeling approaches. While many new general purpose machine learning approaches have emerged, they remain poorly understand and irreconcilable with more traditional statistical modeling approaches. We present a general class of exponential smoothed recurrent neural networks (RNNs) which are well suited to modeling nonstationary dynamical systems arising in industrial applications. In particular, we analyze their capacity to characterize the nonlinear partial autocorrelation structure of time series and directly capture dynamic effects such as seasonality and trends. Application of exponentially smoothed RNNs to forecasting electricity load, weather data, and stock prices highlight the efficacy of exponential smoothing of the hidden state for multistep time series forecasting. The results also suggest that popular, but more complicated neural network architectures originally designed for speech processing are likely over-engineered for industrial forecasting and light-weight exponentially smoothed architectures, trained in a fraction of the time, capture the salient features while being superior and more robust than simple RNNs and autoregressive models. Additionally, uncertainty quantification of Bayesian exponential smoothed RNNs is shown to provide improved coverage.

Title: Depth Selection for Deep ReLU Nets in Feature Extraction and Generalization Abstract: Deep learning is recognized to be capable of discovering deep features for representation learning and pattern recognition without requiring elegant feature engineering techniques by taking advantages of human ingenuity and prior knowledge. Thus it has triggered enormous research activities in machine learning and pattern recognition. One of the most important challenges of deep learning is to fig...

Title: Statistical inference in massive datasets by empirical likelihood Abstract: In this paper, we propose a new statistical inference method for massive data sets, which is very simple and efficient by combining divide-and-conquer method and empirical likelihood. Compared with two popular methods (the bag of little bootstrap and the subsampled double bootstrap), we make full use of data sets, and reduce the computation burden. Extensive numerical studies and real data analysis demonstrate the effectiveness and flexibility of our proposed method. Furthermore, the asymptotic property of our method is derived.

Title: Sizes of simultaneous core partitions Abstract: There is a well-studied correspondence by Jaclyn Anderson between partitions that avoid hooks of length s or t and certain binary strings of length s+t. Using this map, we prove that the total size of a random partition of this kind converges in law to Watson's U2 distribution, as conjectured by Doron Zeilberger.

Title: Forcing axioms and the Galvin number Abstract: We study the Galvin property. We show that various square principles imply that the cofinality of the Galvin number is uncountable (or even greater than $$\aleph _1$$ ). We prove that the proper forcing axiom is consistent with a strong negation of the Glavin property.

Title: Nearest neighbor control for practical stabilization of passive nonlinear systems Abstract: This paper studies static output feedback stabilization of continuous-time (incrementally) passive nonlinear systems where the control actions can only be chosen from a discrete (and possibly finite) set of points. For this purpose, we are working under the assumption that the system under consideration is large-time norm observable and the convex hull of the realizable control actions contains the target constant input (which corresponds to the equilibrium point) in its interior. We propose a nearest-neighbor based static feedback mapping from the output space to the finite set of control actions, that is able to practically stabilize the closed-loop systems. Consequently, we show that for such systems with m-dimensional input space, it is sufficient to have m+1 discrete input points (other than zero for general passive systems or the target constant input for constant-incrementally passive systems). Furthermore, we present a constructive algorithm to design such m+1 nonzero input points that satisfy the conditions for practical stability using our proposed nearest-neighbor control.

Title: Permutation test for the multivariate coefficient of variation in factorial designs Abstract: New inference methods for the multivariate coefficient of variation and its reciprocal, the standardized mean, are presented. While there are various testing procedures for both parameters in the univariate case, it is less known how to do inference in the multivariate setting appropriately. There are some existing procedures but they rely on restrictive assumptions on the underlying distributions. We tackle this problem by applying Wald-type statistics in the context of general, potentially heteroscedastic factorial designs. In addition to the k-sample case, higher-way layouts can be incorporated into this framework allowing the discussion of main and interaction effects. The resulting procedures are shown to be asymptotically valid under the null hypothesis and consistent under general alternatives. To improve the finite sample performance, we suggest permutation versions of the tests and show that the tests’ asymptotic properties can be transferred to them. An exhaustive simulation study compares the new tests, their permutation counterparts and existing methods. To further analyze the differences between the tests, we conduct two illustrative real data examples.

Title: Discounted cost linear quadratic Gaussian control for descriptor systems Abstract: We consider the linear quadratic Gaussian control problem with a discounted cost functional for descriptor systems on the infinite time horizon. Based on recent results from the deterministic framework, we characterise the feasibility of this problem using a linear matrix inequality. In particular, conditions for existence and uniqueness of optimal controls are derived, which are weaker compared to the standard approaches in the literature. We further show that also for the stochastic problem, the optimal control is given in terms of the stabilising solution of the Lur'e equation, which generalises the algebraic Riccati equation. We conclude our paper by examining an application of our theory to fluid dynamics problems.

Title: Embedded pairs for optimal explicit strong stability preserving Runge–Kutta methods Abstract: We construct a family of embedded pairs for optimal explicit strong stability preserving Runge–Kutta methods of order 2≤p≤4 to be used to obtain numerical solution of spatially discretized hyperbolic PDEs. In this construction, the goals include non-defective property, large stability region, and small error values as defined in Dekker and Verwer (1984) and Kennedy et al. (2000). The new family of embedded pairs offer the ability for strong stability preserving (SSP) methods to adapt by varying the step-size. Through several numerical experiments, we assess the overall effectiveness in terms of work versus precision while also taking into consideration accuracy and stability.

Title: Junk news bubbles modelling the rise and fall of attention in online arenas Abstract: In this article, we present a type of media disorder which we call 'junk news bubbles' and which derives from the effort invested by online platforms and their users to identify and circulate contents with rising popularity. Such emphasis on trending matters, we claim, can have two detrimental effects on public debates: first, it shortens the amount of time available to discuss each matter and second, it increases the ephemeral concentration of media attention. We provide a formal description of the dynamic of junk news bubbles, through a mathematical exploration of the famous 'public arenas model' developed by Hilgartner and Bosk in 1988. Our objective is to describe the dynamics of the junk news bubbles as precisely as possible to facilitate its further investigation with empirical data.

Title: Learning to Cache and Caching to Learn: Regret Analysis of Caching Algorithms Abstract: Crucial performance metrics of a caching algorithm include its ability to quickly and accurately learn a popularity distribution of requests. However, a majority of work on analytical performance analysis focuses on hit probability after an asymptotically large time has elapsed. We consider an online learning viewpoint, and characterize the “regret” in terms of the finite time difference between t...

Title: Semantic Private Information Retrieval Abstract: We investigate the problem of semantic private information retrieval (semantic PIR). In semantic PIR, a user retrieves a message out of $K$ independent messages stored in $N$ replicated and non-colluding databases without revealing the identity of the de...

Title: A new smoothing algorithm for jump Markov linear systems Abstract: This paper presents a method for calculating the smoothed state distribution for Jump Markov Linear Systems. More specifically, the paper details a novel two-filter smoother that provides closed-form expressions for the smoothed hybrid state distribution. This distribution can be expressed as a Gaussian mixture with a known, but exponentially increasing, number of Gaussian components as the time index increases. This is accompanied by exponential growth in memory and computational requirements, which rapidly becomes intractable. To ameliorate this, we limit the number of allowed mixture terms by employing a Gaussian likelihood mixture reduction strategy, which results in a computationally tractable, but approximate smoothed distribution. The approximation error can be balanced against computational complexity in order to provide an accurate and practical smoothing algorithm that compares favourably to existing state-of-the-art approaches.

Title: ResNeSt: Split-Attention Networks Abstract: The ability to learn richer network representations generally boosts the performance of deep learning models. To improve representation-learning in convolutional neural networks, we present a multi-branch architecture, which applies channel-wise attention across different network branches to leverage the complementary strengths of both feature-map attention and multi-path representation. Our proposed Split-Attention module provides a simple and modular computation block that can serve as a drop-in replacement for the popular residual block, while producing more diverse representations via cross-feature interactions. Adding a Split-Attention module into the architecture design space of RegNet-Y and FBNetV2 directly improves the performance of the resulting network. Replacing residual blocks with our Split-Attention module, we further design a new variant of the ResNet model, named ResNeSt, which outperforms EfficientNet in terms of the accuracy/latency trade-off.

Title: A Stochastic-Robust Approach for Resilient Microgrid Investment Planning Under Static and Transient Islanding Security Constraints Abstract: When planning the investment in Microgrids (MGs), usually static security constraints are included to ensure their resilience and ability to operate in islanded mode. However, unscheduled islanding events may trigger cascading disconnections of Distributed Energy Resources (DERs) inside the MG due to the transient response, leading to a partial or full loss of load. In this paper, a min-max-min, h...

Title: Modeling and mitigating human annotation errors to design efficient stream processing systems with human-in-the-loop machine learning Abstract: •Study of human annotation task in hybrid stream processing systems.•Presenting a generic human error framework of serial ordering-based mistakes and slips.•Verifying of the proposed human error framework through extensive experiments.•Presenting a novel method for human error-mitigation in an active learning paradigm.•Validating the novel method through simulation-based experiments.

Title: Cartesian Lattice Counting by the Vertical 2-sum Abstract: A vertical 2-sum of a two-coatom lattice L and a two-atom lattice U is obtained by removing the top of L and the bottom of U, and identifying the coatoms of L with the atoms of U. This operation creates one or two nonisomorphic lattices depending on the symmetry case. Here the symmetry cases are analyzed, and a recurrence relation is presented that expresses the number of nonisomorphic vertical 2-sums in some desired family of graded lattices. Nonisomorphic, vertically indecomposable modular and distributive lattices are counted and classified up to 35 and 60 elements respectively. Asymptotically their numbers are shown to be at least Ω(2.3122n) and Ω(1.7250n), where n is the number of elements. The number of semimodular lattices is shown to grow faster than any exponential in n.

Title: Innovative and Additive Outlier Robust Kalman Filtering With a Robust Particle Filter Abstract: In this paper, we propose CE-BASS, a particle mixture Kalman filter which is robust to both innovative and additive outliers, and able to fully capture multi-modality in the distribution of the hidden state. Furthermore, the particle sampling approach re-samples past states, which enables CE-BASS to handle innovative outliers which are not immediately visible in the observations, such as trend cha...

Title: Full Duplex Integrated Access and Backhaul for 5G NR: Analyses and Prototype Measurements Abstract: Researchers for the third-generation partnership project (3GPP) have been exploring — as a cost-effective alternative to wired backhaul — integrated access and backhaul (IAB) frameworks for 5G new radio (NR). A promising solution for this framework is the integration of full duplex (FD) technologies to enhance the spectral efficiency and efficiently utilize network resources. This approach, which we refer to as FD IAB, involves a significant technical challenge — self-interference (SI) in the IAB framework. In fact, this challenge casts doubt over the performance and feasibility of FD IAB. In this article, we introduce the FD IAB framework and its enabling technologies and also evaluate the framework's link-level SI reduction and system-level downlink throughput performance. Thereafter, we validate the attenuation level in the SI channel with antenna separation and high directional gain through 28 GHz hardware prototype measurements. Our numerical evaluations and hardware prototype measurements confirm that FD IAB represents a promising framework for 5G NR.

Title: A Discrete Convex Min-Max Formula for Box-TDI Polyhedra Abstract: A min-max formula is proved for the minimum of an integer-valued separable discrete convex function where the minimum is taken over the set of integral elements of a box total dual integral (box-TDI) polyhedron. One variant of the theorem uses the notion of conjugate function (a fundamental concept in non-linear optimization) but we also provide another version that avoids conjugates, and its spirit is conceptually closer to the standard form of classic min-max theorems in combinatorial optimization. The presented framework provides a unified background for separable convex minimization over the set of integral elements of the intersection of two integral base-polyhedra, submodular flows, L-convex sets, and polyhedra defined by totally unimodular (TU) matrices. As an application, we show how inverse combinatorial optimization problems can be covered by this new framework.

Title: A large English–Thai parallel corpus from the web and machine-generated text Abstract: The primary objective of our work is to build a large-scale English–Thai dataset for training neural machine translation models. We construct scb-mt-en-th-2020, an English–Thai machine translation dataset with over 1 million segment pairs, curated from various sources: news, Wikipedia articles, SMS messages, task-based dialogs, web-crawled data, government documents, and text artificially generated by a pretrained language model. We present the methods for gathering data, aligning texts, and removing preprocessing noise and translation errors automatically. We also train machine translation models based on this dataset to assess the quality of the corpus. Our models perform comparably to Google Translation API (as of May 2020) for Thai–English and outperform Google when the Open Parallel Corpus (OPUS) is included in the training data for both Thai–English and English–Thai translation. The dataset is available for public use under CC-BY-SA 4.0 License. The pre-trained models and source code to reproduce our work are available under Apache-2.0 License.

Title: LONGER LIFESPAN FOR MANY SOLUTIONS OF THE KIRCHHOFF EQUATION Abstract: We consider the Kirchhoff equation partial derivative(tt) u - Delta u(1+ integral(Td) vertical bar del u vertical bar(2)) = 0 on the d-dimensional torus T-d, and its Cauchy problem with initial data u(0, x), partial derivative(t)u(0, x) of size epsilon in the Sobolev class. The effective equation for the dynamics at the quintic order, obtained in previous papers by quasilinear normal form, contains resonances corresponding to nontrivial terms in the energy estimates. Such resonances cannot be avoided by tuning external parameters (simply because the Kirchhoff equation does not contain parameters). In this paper we introduce nonresonance conditions on the initial data of the Cauchy problem and prove a lower bound epsilon(-6) for the lifespan of the corresponding solutions (the standard local theory gives epsilon(-2), and the normal form for the cubic terms gives epsilon(-4)). The proof relies on the fact that, under these nonresonance conditions, the growth rate of the "superactions" of the effective equations on large time intervals is smaller (by a factor epsilon(-2)) than its a priori estimate based on the normal form for the cubic terms. The set of initial data satisfying such nonresonance conditions contains several nontrivial examples that are discussed in the paper.

Title: RGCF: Refined graph convolution collaborative filtering with concise and expressive embedding Abstract: Graph Convolution Networks (GCNs) have attracted significant attention and have become the most popular method for learning graph representations. In recent years, many efforts have focused on integrating GCNs into recommender tasks and have made remarkable progress. At its core is to explicitly capture the high-order connectivities between nodes in the user-item bipartite graph. However, we found some potential drawbacks existed in the traditional GCN-based recommendation models are that the excessive information redundancy yield by the nonlinear graph convolution operation reduces the expressiveness of the resultant embeddings, and the important popularity features that are effective in sparse recommendation scenarios are not encoded in the embedding generation process. In this work, we develop a novel GCN-based recommendation model, named Refined Graph convolution Collaborative Filtering (RGCF), where a refined graph convolution structure is designed to match non-semantic ID inputs. In addition, a new fine-tuned symmetric normalization is proposed to mine node popularity characteristics and further incorporate the popularity features into the embedding learning process. Extensive experiments were conducted on three public million-size datasets, and the RGCF improved by an average of 13.45% over the state-of-the-art baseline. Further comparative experiments validated the effectiveness and rationality of each part of our proposed RGCF. We released our code at https://github.com/hfutmars/RGCF.

Title: Algorithm 1019: A Task-based Multi-shift QR/QZ Algorithm with Aggressive Early Deflation Abstract: AbstractThe QR algorithm is one of the three phases in the process of computing the eigenvalues and the eigenvectors of a dense nonsymmetric matrix. This paper describes a task-based QR algorithm for reducing an upper Hessenberg matrix to real Schur form. The task-based algorithm also supports generalized eigenvalue problems (QZ algorithm) but this paper concentrates on the standard case. The task-based algorithm adopts previous algorithmic improvements, such as tightly-coupled multi-shifts and Aggressive Early Deflation (AED), and also incorporates several new ideas that significantly improve the performance. This includes, but is not limited to, the elimination of several synchronization points, the dynamic merging of previously separate computational steps, the shortening and the prioritization of the critical path, and experimental GPU support. The task-based implementation is demonstrated to be multiple times faster than multi-threaded LAPACK and ScaLAPACK in both single-node and multi-node configurations on two different machines based on Intel and AMD CPUs. The implementation is built on top of the StarPU runtime system and is part of the open-source StarNEig library.

Title: Re-Thinking Co-Salient Object Detection Abstract: In this article, we conduct a comprehensive study on the co-salient object detection (CoSOD) problem for images. CoSOD is an emerging and rapidly growing extension of salient object detection (SOD), which aims to detect the co-occurring salient objects in a group of images. However, existing CoSOD datasets often have a serious data bias, assuming that each group of images contains salient objects of similar visual appearances. This bias can lead to the ideal settings and effectiveness of models trained on existing datasets, being impaired in real-life situations, where similarities are usually semantic or conceptual. To tackle this issue, we first introduce a new benchmark, called CoSOD3k in the wild, which requires a large amount of semantic context, making it more challenging than existing CoSOD datasets. Our CoSOD3k consists of 3,316 high-quality, elaborately selected images divided into 160 groups with hierarchical annotations. The images span a wide range of categories, shapes, object sizes, and backgrounds. Second, we integrate the existing SOD techniques to build a unified, trainable CoSOD framework, which is long overdue in this field. Specifically, we propose a novel CoEG-Net that augments our prior model EGNet with a co-attention projection strategy to enable fast common information learning. CoEG-Net fully leverages previous large-scale SOD datasets and significantly improves the model scalability and stability. Third, we comprehensively summarize 40 cutting-edge algorithms, benchmarking 18 of them over three challenging CoSOD datasets (iCoSeg, CoSal2015, and our CoSOD3k), and reporting more detailed (i.e., group-level) performance analysis. Finally, we discuss the challenges and future works of CoSOD. We hope that our study will give a strong boost to growth in the CoSOD community. The benchmark toolbox and results are available on our project page at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://dpfan.net/CoSOD3K</uri> .

Title: Monitoring Browsing Behavior of Customers in Retail Stores via RFID Imaging Abstract: In this paper, we propose to use commercial off-the-shelf (COTS) <i>monostatic</i> RFID devices (i.e. which use a single antenna at a time for both transmitting and receiving RFID signals to and from the tags) to monitor browsing activity of customers in front of display items in places such as retail stores. To this end, we propose <i>TagSee</i> , a multi-person imaging system based on monostatic RFID imaging. TagSee is based on the insight that when customers are browsing the items on a shelf, they stand between the tags deployed along the boundaries of the shelf and the reader, which changes the multi-paths that the RFID signals travel along, and both the RSS and phase values of the RFID signals that the reader receives change. Based on these variations observed by the reader, TagSee constructs a coarse grained image of the customers. Afterwards, TagSee identifies the items that are being browsed by the customers by analyzing the constructed images. The key novelty of this paper is on achieving browsing behavior monitoring of multiple customers in front of display items by constructing coarse grained images via robust, analytical model-driven deep learning based, RFID imaging. To achieve this, we first mathematically formulate the problem of imaging humans using monostatic RFID devices and derive an approximate analytical imaging model that correlates the variations caused by human obstructions in the RFID signals. Based on this model, we then develop a deep learning framework to robustly image customers with high accuracy. We implement TagSee scheme using a Impinj Speedway R420 reader and SMARTRAC DogBone RFID tags. TagSee can achieve a TPR of more than <inline-formula><tex-math notation="LaTeX">${\sim }90\%$</tex-math></inline-formula> and a FPR of less than <inline-formula><tex-math notation="LaTeX">${\sim }10\%$</tex-math></inline-formula> in multi-person scenarios using training data from just 3-4 users.

Title: A Sylvester-Gallai Result for Concurrent Lines in the Complex Plane Abstract: We show that if a finite non-collinear set of points in \n$$\\mathbb {C}^2$$\n\n lies on a family of m concurrent lines, and if one of those lines contains more than \n$$m-2$$\n\n points, there exists a line passing through exactly two points of the set. The bound \n$$m-2$$\n\n in our result is optimal. Our main theorem resolves a conjecture of Frank de Zeeuw, and generalizes a result of Kelly and Nwankpa.

Title: Human Trajectory Forecasting in Crowds: A Deep Learning Perspective Abstract: Since the past few decades, human trajectory forecasting has been a field of active research owing to its numerous real-world applications: evacuation situation analysis, deployment of intelligent transport systems, traffic operations, to name a few. In this work, we cast the problem of human trajectory forecasting as learning a representation of human social interactions. Early works handcrafted this representation based on domain knowledge. However, social interactions in crowded environments are not only diverse but often subtle. Recently, deep learning methods have outperformed their handcrafted counterparts, as they learn about human-human interactions in a more generic data-driven fashion. In this work, we present an in-depth analysis of existing deep learning-based methods for modelling social interactions. We propose two domain-knowledge inspired data-driven methods to effectively capture these social interactions. To objectively compare the performance of these interaction-based forecasting models, we develop a large scale interaction-centric benchmark TrajNet++ a significant yet missing component in the field of human trajectory forecasting. We propose novel performance metrics that evaluate the ability of a model to output socially acceptable trajectories. Experiments on TrajNet++ validate the need for our proposed metrics, and our method outperforms competitive baselines on both real-world and synthetic datasets.

Title: Distributed Formation Navigation of Constrained Second-Order Multiagent Systems With Collision Avoidance and Connectivity Maintenance Abstract: In this article, we consider the distributed formation navigation problem of second-order multiagent systems subject to both velocity and input constraints. Both collision avoidance and connectivity maintenance of the network are considered in the controller design. A control barrier function method is employed to achieve multiple control objectives simultaneously while satisfying the velocity and...

Title: Spatio-Temporal Meta Learning for Urban Traffic Prediction Abstract: Predicting urban traffic is of great importance to intelligent transportation systems and public safety, yet is very challenging in three aspects: 1) complex spatio-temporal correlations of urban traffic, including spatial correlations between locations along with temporal correlations among timestamps; 2) spatial diversity of such spatio-temporal correlations, which varies from location to location and depends on the surrounding geographical information, e.g., points of interests and road networks; and 3) temporal diversity of such spatio-temporal correlations, which is highly influenced by dynamic traffic states. To tackle these challenges, we proposed a deep meta learning based model, entitled ST-MetaNet <inline-formula><tex-math notation="LaTeX">$^+$</tex-math></inline-formula> , to <i>collectively</i> predict traffic in all locations at the same time. ST-MetaNet <inline-formula><tex-math notation="LaTeX">$^+$</tex-math></inline-formula> employs a sequence-to-sequence architecture, consisting of an encoder to learn historical information and a decoder to make predictions step by step. Specifically, the encoder and decoder have the same network structure, consisting of meta graph attention networks and meta recurrent neural networks, to capture diverse spatial and temporal correlations, respectively. Furthermore, the weights (parameters) of meta graph attention networks and meta recurrent neural networks are generated from the embeddings of geo-graph attributes and the traffic context learned from dynamic traffic states. Extensive experiments were conducted based on three real-world datasets to illustrate the effectiveness of ST-MetaNet <inline-formula><tex-math notation="LaTeX">$^+$</tex-math></inline-formula> beyond several state-of-the-art methods.

Title: Web Service QoS Prediction via Collaborative Filtering: A Survey Abstract: With the growing number of competing Web services that provide similar functionality, Quality-of-Service (QoS) prediction is becoming increasingly important for various QoS-aware approaches of Web services. Collaborative filtering (CF), which is among the most successful personalized prediction techniques for recommender systems, has been widely applied to Web service QoS prediction. In addition to using conventional CF techniques, a number of studies extend the CF approach by incorporating additional information about services and users, such as location, time, and other contextual information from the service invocations. There are also some studies that address other challenges in QoS prediction, such as adaptability, credibility, privacy preservation, and so on. In this survey, we summarize and analyze the state-of-the-art CF QoS prediction approaches of Web services and discuss their features and differences. We also present several Web service QoS datasets that have been used as benchmarks for evaluating the predition accuracy and outline some possible future research directions.

Title: SofGAN: A Portrait Image Generator with Dynamic Styling Abstract: AbstractRecently, Generative Adversarial Networks (GANs) have been widely used for portrait image generation. However, in the latent space learned by GANs, different attributes, such as pose, shape, and texture style, are generally entangled, making the explicit control of specific attributes difficult. To address this issue, we propose a SofGAN image generator to decouple the latent space of portraits into two subspaces: a geometry space and a texture space. The latent codes sampled from the two subspaces are fed to two network branches separately, one to generate the 3D geometry of portraits with canonical pose, and the other to generate textures. The aligned 3D geometries also come with semantic part segmentation, encoded as a semantic occupancy field (SOF). The SOF allows the rendering of consistent 2D semantic segmentation maps at arbitrary views, which are then fused with the generated texturemaps and stylized to a portrait photo using our semantic instance-wise module. Through extensive experiments, we show that our system can generate high-quality portrait images with independently controllable geometry and texture attributes. The method also generalizes well in various applications, such as appearance-consistent facial animation and dynamic styling.

Title: Fully discrete loosely coupled Robin-Robin scheme for incompressible fluid–structure interaction: stability and error analysis Abstract: We consider a fully discrete loosely coupled scheme for incompressible fluid–structure interaction based on the time semi-discrete splitting method introduced in Burman et al. (Numer Methods Partial Differ Equ, 2021). The splittling method uses a Robin-Robin type coupling that allows for a segregated solution of the solid and the fluid systems, without inner iterations. For the discretisation in space we consider piecewise affine continuous finite elements for all the fields and ensure the inf-sup condition by using a Brezzi-Pitkäranta type pressure stabilization. The interfacial fluid-stresses are evaluated in a variationally consistent fashion, that is shown to admit an equivalent Lagrange multiplier formulation. We prove that the method is unconditionally stable and robust with respect to the amount of added-mass in the system. Furthermore, we provide an error estimate that shows the error in the natural energy norm for the system is $${\mathcal {O}}\big (\sqrt{T}(\sqrt{\Delta t} + h)\big )$$ where T is the final time, $$\Delta t$$ the time-step length and h the space discretization parameter.

Title: PaMIR: Parametric Model-Conditioned Implicit Representation for Image-Based Human Reconstruction Abstract: Modeling 3D humans accurately and robustly from a single image is very challenging, and the key for such an ill-posed problem is the 3D representation of the human models. To overcome the limitations of regular 3D representations, we propose Parametric Model-Conditioned Implicit Representation (PaMIR), which combines the parametric body model with the free-form deep imp...

Title: Multi-resolution beta-divergence NMF for blind spectral unmixing Abstract: •We formulate a multi-resolution NMF problem for any β-divergence.•We propose a novel algorithm based on the multiplicative updates (MU).•We apply, for the first time, such a model to the blind unmixing of audio spectrograms.•We apply our model for the hyper/multi-spectral image super-resolution problem.•Our MU outperform state-of-the-art algorithms in the presence of non-Gaussian noise.

Title: A technique for obtaining true approximations for k-center with covering constraints Abstract: There has been a recent surge of interest in incorporating fairness aspects into classical clustering problems. Two recently introduced variants of the k-Center problem in this spirit are Colorful k-Center, introduced by Bandyapadhyay, Inamdar, Pai, and Varadarajan, and lottery models, such as the Fair Robust k-Center problem introduced by Harris, Pensyl, Srinivasan, and Trinh. To address fairness aspects, these models, compared to traditional k-Center, include additional covering constraints. Prior approximation results for these models require to relax some of the normally hard constraints, like the number of centers to be opened or the involved covering constraints, and therefore, only obtain constant-factor pseudo-approximations. In this paper, we introduce a new approach to deal with such covering constraints that leads to (true) approximations, including a 4-approximation for Colorful k-Center with constantly many colors—settling an open question raised by Bandyapadhyay, Inamdar, Pai, and Varadarajan—and a 4-approximation for Fair Robust k-Center, for which the existence of a (true) constant-factor approximation was also open. We complement our results by showing that if one allows an unbounded number of colors, then Colorful k-Center admits no approximation algorithm with finite approximation guarantee, assuming that $$\mathtt {P}\ne \mathtt {NP}$$ . Moreover, under the Exponential Time Hypothesis, the problem is inapproximable if the number of colors grows faster than logarithmic in the size of the ground set.

Title: Secure Distributed Matrix Computation With Discrete Fourier Transform Abstract: We consider the problem of secure distributed matrix computation (SDMC), where a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">user</i> queries a function of data matrices generated at distributed <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">source</i> nodes. We assume the availability of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$N$ </tex-math></inline-formula> honest but curious computation servers, which are connected to the sources, the user, and each other through orthogonal and reliable communication links. Our goal is to minimize the amount of data that must be transmitted from the sources to the servers, called the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">upload cost</i> , while guaranteeing that no <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$T$ </tex-math></inline-formula> colluding servers can learn any information about the source matrices, and the user cannot learn any information beyond the computation result. We first focus on secure distributed matrix multiplication (SDMM), considering two matrices, and propose a novel polynomial coding scheme using the properties of finite field discrete Fourier transform, which achieves an upload cost significantly lower than the existing results in the literature. We then generalize the proposed scheme to include straggler mitigation, and to the multiplication of multiple matrices while keeping the input matrices, the intermediate computation results, as well as the final result secure against any <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$T$ </tex-math></inline-formula> colluding servers. We also consider a special case, called computation with own data, where the data matrices used for computation belong to the user. In this case, we drop the security requirement against the user, and show that the proposed scheme achieves the minimal upload cost. We then propose methods for performing other common matrix computations securely on distributed servers, including changing the parameters of secret sharing, matrix transpose, matrix exponentiation, solving a linear system, and matrix inversion, which are then used to show how arbitrary matrix polynomials can be computed securely on distributed servers using the proposed procedure.

Title: Dynamic Social Learning Under Graph Constraints Abstract: We introduce a model of graph-constrained dynamic choice with reinforcement modeled by positively <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\alpha$</tex-math></inline-formula> -homogeneous rewards. We show that its empirical process, which can be written as a stochastic approximation recursion with Markov noise, has the same probability law as a certain vertex reinforced random walk. We use this equivalence to show that for <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\alpha &gt; 0$</tex-math></inline-formula> , the asymptotic outcome concentrates around the optimum in a certain limiting sense when “annealed” by letting <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\alpha \uparrow \infty$</tex-math></inline-formula> slowly.

Title: A characterization of 2-threshold functions via pairs of prime segments Abstract: A {0,1}-valued function on a two-dimensional rectangular grid is called threshold if its sets of zeros and ones are separable by a straight line. In this paper we study 2-threshold functions, i.e. functions representable as the conjunction of two threshold functions. We provide a characterization of 2-threshold functions by pairs of oriented prime segments, where each such segment is defined by an ordered pair of adjacent integer points.

Title: Waypoint routing on bounded treewidth graphs Abstract: •In the Waypoint Routing Problem we are given a capacitated and weighted graph G, a set W⊆V(G), and vertices s,t∈V(G).•The goal is to find s,t-walk of minimum weight respecting edge capacities and visiting all waypoints w∈W.•We show that the problem is in FPT complexity class with respect to the treewidth and present algorithm working in 2O(tw)⋅n time.•In addition, we prove that the running time of our algorithm is optimal for the problem under the Exponential Time Hypothesis.

Title: BS4NN: Binarized Spiking Neural Networks with Temporal Coding and Learning Abstract: We recently proposed the S4NN algorithm, essentially an adaptation of backpropagation to multilayer spiking neural networks that use simple non-leaky integrate-and-fire neurons and a form of temporal coding known as time-to-first-spike coding. With this coding scheme, neurons fire at most once per stimulus, but the firing order carries information. Here, we introduce BS4NN, a modification of S4NN in which the synaptic weights are constrained to be binary ( $$+$$ 1 or − 1), in order to decrease memory (ideally, one bit per synapse) and computation footprints. This was done using two sets of weights: firstly, real-valued weights, updated by gradient descent, and used in the backward pass of backpropagation, and secondly, their signs, used in the forward pass. Similar strategies have been used to train (non-spiking) binarized neural networks. The main difference is that BS4NN operates in the time domain: spikes are propagated sequentially, and different neurons may reach their threshold at different times, which increases computational power. We validated BS4NN on two popular benchmarks, MNIST and Fashion-MNIST, and obtained reasonable accuracies for this sort of network (97.0% and 87.3% respectively) with a negligible accuracy drop with respect to real-valued weights (0.4% and 0.7%, respectively). We also demonstrated that BS4NN outperforms a simple BNN with the same architectures on those two datasets (by 0.2% and 0.9% respectively), presumably because it leverages the temporal dimension.

Title: Optimizing Information Freshness via Multiuser Scheduling With Adaptive NOMA/OMA Abstract: This paper considers a wireless network with a base station (BS) conducting timely status updates to multiple clients via adaptive non-orthogonal multiple access (NOMA)/orthogonal multiple access (OMA). Specifically, the BS is able to adaptively switch between NOMA and OMA for the downlink transmission to optimize the information freshness of the network, characterized by the Age of Information (A...

Title: Viscosity solutions to first order path-dependent Hamilton–Jacobi–Bellman equations in Hilbert space Abstract: In this article, a notion of viscosity solutions is introduced for first order path-dependent Hamilton–Jacobi–Bellman (PHJB) equations associated with optimal control problems for path-dependent evolution equations in Hilbert space. We identify the value functional of optimal control problems as unique viscosity solution to the associated PHJB equations without the assumption (A.4) on page 231 of Li and Yong (1995). We also show that our notion of viscosity solutions is consistent with the corresponding notion of classical solutions, and satisfies a stability property.

Title: Green-PoW: An energy-efficient blockchain Proof-of-Work consensus algorithm Abstract: This paper opts to mitigate the energy-inefficiency of the Blockchain Proof-of-Work (PoW) consensus algorithm by rationally repurposing the power spent during the mining process. The original PoW mining scheme is designed to consider one block at a time and assign a reward to the first place winner of a computation race. To reduce the mining-related energy consumption, we propose to compensate the computation effort of the runner(s)-up of a mining round, by granting them exclusivity of solving the upcoming block in the next round. This will considerably reduce the number of competing nodes in the next round and consequently, the consumed energy. Our proposed scheme divides time into epochs, where each comprises two mining rounds; in the first one, all network nodes can participate in the mining process, whereas in the second round only runners-up can take part. Thus, the overall mining energy consumption can be reduced to nearly 50%. To the best of our knowledge, our proposed scheme is the first to considerably decrease the energy consumption of the original PoW algorithm. Our analysis demonstrates the effectiveness of our scheme in reducing energy consumption, the probability of fork occurrences, the level of mining centralization presented in the original PoW algorithm, and the effect of transaction censorship attack.

Title: The uncertainty principle over finite fields Abstract: In this paper we study the uncertainty principle (UP) connecting a function over a finite field and its Mattson-Solomon polynomial, which is a kind of Fourier transform in positive characteristic. Three versions of the UP over finite fields are studied, in connection with the asymptotic theory of cyclic codes. We first show that no finite field satisfies the strong version of UP, introduced recently by Evra, Kowalsky, Lubotzky, 2017. A refinement of the weak version is given, by using the asymptotic Plotkin bound. A naive version, which is the direct analogue over finite fields of the Donoho-Stark bound over the complex numbers, is proved by using the BCH bound. It is strong enough to show that there exist sequences of cyclic codes of length n, arbitrary rate, and minimum distance Omega(n(alpha)) for all 0 < alpha < 1/2. Finally, a connection with Ramsey Theory is pointed out. (c) 2021 Elsevier B.V. All rights reserved.

Title: A Novel BGCapsule Network for Text Classification. Abstract: Several text classification tasks such as sentiment analysis, news categorization, multi-label classification and opinion classification are challenging problems even for modern deep learning networks. Recently, Capsule Networks (CapsNets) are proposed for image classification. It has been shown that CapsNets have several advantages over Convolutional Neural Networks (CNNs), while their validity in the domain of text has been less explored. In this paper, we propose a novel hybrid architecture viz., BGCapsule, which is a Capsule model preceded by an ensemble of Bidirectional Gated Recurrent Units (BiGRU) for several text classification tasks. We employed an ensemble of Bidirectional GRUs for feature extraction layer preceding the primary capsule layer. The hybrid architecture, after performing basic pre-processing steps, consists of five layers: an embedding layer based on GloVe, a BiGRU based ensemble layer, a primary capsule layer, a flatten layer and fully connected ReLU layer followed by a fully connected softmax layer. In order to evaluate the effectiveness of BGCapsule, we conducted extensive experiments on five benchmark datasets (ranging from 10,000 records to 700,000 records) including Movie Review (MR Imdb 2005), AG News dataset, Dbpedia ontology dataset, Yelp Review Full dataset and Yelp review polarity dataset. These benchmarks cover several text classification tasks such as news categorization, sentiment analysis, multiclass classification, multi-label classification and opinion classification. We found that our proposed architecture (BGCapsule) achieves better accuracy compared to the existing methods without the help of any external linguistic knowledge such as positive sentiment keywords and negative sentiment keywords. Further, BGCapsule converged faster compared to other extant techniques.

Title: Long term optimal investment with regime switching: inflation, information and short sales Abstract: Financial models are based on the standard assumptions of frictionless markets, complete information, no transaction costs and no taxes and borrowing and short selling without restrictions. Short-selling bans around the world after the global financial crisis and in several exchanges during the COVID 19 period, become more and more important. This paper bridges the gap by providing for the first time in the literature a model that accounting explicitly and simultaneously for inflation, information costs and short sales in the portfolio performance with regime switching. Our model can be used by portfolio managers to assess the impact of these market imperfections on portfolio decisions.

Title: A statistical and deep learning-based daily infected count prediction system for the coronavirus pandemic Abstract: We present new data analytics-based predictions results that can help governments to plan their future actions and also help medical services to be better prepared for the future. Our system can predict new corona cases with 99.82% accuracy using susceptible infected recovered (SIR) model. We have predicted the results of new COVID cases per day for dense and highly populated country i.e. India. We found that traditional statistical methods will not work efficiently as they do not consider the limited population in a particular country. Using the data analytics-based curve we predicted four most likely possibilities for the number of new cases in India. Hence, we expect that the results mentioned in the manuscript help people to better understand the progress of this disease.

Title: Monitoring and Tracking the Evolution of a Viral Epidemic Through Nonlinear Kalman Filtering: Application to the COVID-19 Case Abstract: This work presents a novel methodology for systematically processing the time series that report the number of positive, recovered and deceased cases from a viral epidemic, such as Covid-19. The main objective is to unveil the evolution of the number of real infected people, and consequently to predict the peak of the epidemic and subsequent evolution. For this purpose, an original nonlinear model...

Title: Average Top-k Aggregate Loss for Supervised Learning Abstract: In this work, we introduce the average top-$k$k ($\mathrm {AT...

Title: The UU-Net: Reversible Face De-Identification for Visual Surveillance Video Footage Abstract: We propose a reversible face de-identification method for video surveillance data, where landmark-based techniques cannot be reliably used. Our solution generates a photorealistic de-identified stream that meets the data protection regulations and can be publicly released under minimal privacy concerns. Notably, such stream still encapsulates the information required to later reconstruct the original scene, which is useful for scenarios, such as crime investigation, where subjects identification is of most importance. Our learning process jointly optimizes two main components: 1) a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">public</i> module, that receives the raw data and generates the de-identified stream; and 2) a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">private</i> module, designed for security authorities, that receives the public stream and reconstructs the original data, disclosing the actual IDs of the subjects in a scene. The proposed solution is landmarks-free and uses a conditional generative adversarial network to obtain synthetic faces that preserve pose, lighting, background information and even facial expressions. Also, we keep full control over the set of soft facial attributes to be preserved/changed between the raw/de-identified data, which extends the range of applications for the proposed solution. Our experiments were conducted in three visual surveillance datasets (BIODI, MARS and P-DESTRE) plus one video face data set (YouTube Faces), showing highly encouraging results. The source code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/hugomcp/uu-net</uri> .

Title: Tuza's conjecture for random graphs Abstract: A celebrated conjecture of Tuza says that in any (finite) graph, the minimum size of a cover of triangles by edges is at most twice the maximum size of a set of edge-disjoint triangles. Resolving a recent question of Bennett, Dudek, and Zerbib, we show that this is true for random graphs; more precisely: for any p = p(n), P(G(n,p) satisfies Tuza's conjecture) -> 1(as n -> infinity).

Title: Robust Geodesic Regression Abstract: This paper studies robust regression for data on Riemannian manifolds. Geodesic regression is the generalization of linear regression to a setting with a manifold-valued dependent variable and one or more real-valued independent variables. The existing work on geodesic regression uses the sum-of-squared errors to find the solution, but as in the classical Euclidean case, the least-squares method is highly sensitive to outliers. In this paper, we use M-type estimators, including the L-1, Huber and Tukey biweight estimators, to perform robust geodesic regression, and describe how to calculate the tuning parameters for the latter two. We show that, on compact symmetric spaces, all M-type estimators are maximum likelihood estimators, and argue in favor of a general preference for the L-1 estimator over the L-2 and Huber estimators on high-dimensional spaces. A derivation of the Riemannian normal distribution on S-n and H-n is also included. Results from numerical examples, including analysis of real neuroimaging data, demonstrate the promising empirical properties of the proposed approach.

Title: On weak G-completeness for fuzzy metric spaces Abstract: In this paper, we provide equivalent characterizations of weak G-complete fuzzy metric spaces. Since such spaces are complete, we also characterize fuzzy metric spaces that have weak G-complete fuzzy metric completions. Moreover, we establish analogous results for classical metric spaces.

Title: Automatic personality prediction: an enhanced method using ensemble modeling Abstract: Human personality is significantly represented by those words which he/she uses in his/her speech or writing. As a consequence of spreading the information infrastructures (specifically the Internet and social media), human communications have reformed notably from face to face communication. Generally, Automatic Personality Prediction (or Perception) (APP) is the automated forecasting of the personality on different types of human generated/exchanged contents (like text, speech, image, video, etc.). The major objective of this study is to enhance the accuracy of APP from the text. To this end, we suggest five new APP methods including term frequency vector-based, ontology-based, enriched ontology-based, latent semantic analysis (LSA)-based, and deep learning-based (BiLSTM) methods. These methods as the base ones, contribute to each other to enhance the APP accuracy through ensemble modeling (stacking) based on a hierarchical attention network (HAN) as the meta-model. The results show that ensemble modeling enhances the accuracy of APP.

Title: DISCO PAL: Diachronic Spanish sonnet corpus with psychological and affective labels Abstract: Nowadays, there are many applications of text mining over corpora from different languages. However, most of them are based on texts in prose, lacking applications that work with poetry texts. An example of an application of text mining in poetry is the usage of features derived from their individual words in order to capture the lexical, sublexical and interlexical meaning, and infer the General Affective Meaning (GAM) of the text. However, even though this proposal has been proved as useful for poetry in some languages, there is a lack of studies for both Spanish poetry and for highly-structured poetic compositions such as sonnets. This article presents a study over an annotated corpus of Spanish sonnets, in order to analyse if it is possible to build features from their individual words for predicting their GAM. The purpose of this is to model sonnets at an affective level. The article also analyses the relationship between the GAM of the sonnets and the content itself. For this, we consider the content from a psychological perspective, identifying with tags when a sonnet is related to a specific term. Then, we study how GAM changes according to each of those psychological terms. The corpus used contains 274 Spanish sonnets from authors of different centuries, from fifteenth to nineteenth. This corpus was annotated by different domain experts. The experts annotated the poems with affective and lexico-semantic features, as well as with domain concepts that belong to psychology. Thanks to this, the corpus of sonnets can be used in different applications, such as poetry recommender systems, personality text mining studies of the authors, or the usage of poetry for therapeutic purposes.

Title: Geometric bounds for convergence rates of averaging algorithms Abstract: We develop a generic method for bounding the convergence rate of an averaging algorithm running in a multiagent system with a time-varying network, where the associated stochastic matrices have a time-independent Perron vector. The resulting bounds depend on geometric parameters of the dynamic communication graph such as the weighted diameter or the bottleneck measure.

Title: Character sums over affine spaces and applications Abstract: Given a finite field Fq, a positive integer n and an Fq-affine space A⊆Fqn, we show that a result of Katz provides a new bound on the sum ∑a∈Aχ(a), where χ is a multiplicative character of Fqn. The paper is focused on the applicability of this estimate to results regarding the existence of special primitive elements in Fqn. In particular, we obtain substantial improvements on previous works.

Title: Vortex Filament Equation for a Regular Polygon in the Hyperbolic Plane Abstract: The aim of this paper is twofold. First, we show the evolution of the vortex filament equation (VFE) for a regular planar polygon in the hyperbolic space. Unlike in the Euclidean space, the planar polygon is open and both of its ends grow up exponentially, which makes the problem more challenging from a numerical point of view. However, using a finite difference scheme in space combined with a fourth-order Runge–Kutta method in time and fixed boundary conditions, we show that the numerical solution is in complete agreement with the one obtained by means of algebraic techniques. Second, as in the Euclidean case, we claim that, at infinitesimal times, the evolution of VFE for a planar polygon as the initial datum can be described as a superposition of several one-corner initial data. As a consequence, not only can we compute the speed of the center of mass of the planar polygon, but the relationship also allows us to compare the time evolution of any of its corners with the evolution in the Euclidean case.

Title: Event-Driven Off-Policy Reinforcement Learning for Control of Interconnected Systems Abstract: In this article, we introduce a novel approximate optimal decentralized control scheme for uncertain input-affine nonlinear-interconnected systems. In the proposed scheme, we design a controller and an event-triggering mechanism (ETM) at each subsystem to optimize a local performance index and reduce redundant control updates, respectively. To this end, we formulate a noncooperative dynamic game a...

Title: Modified BBO-Based Multivariate Time-Series Prediction System With Feature Subset Selection and Model Parameter Optimization Abstract: Multivariate time-series prediction is a challenging research topic in the field of time-series analysis and modeling, and is continually under research. The echo state network (ESN), a type of efficient recurrent neural network, has been widely used in time-series prediction, but when using ESN, two crucial problems have to be confronted: 1) how to select the optimal subset of input features and ...

Title: Output Tracking Control of Single-Input–Multioutput Systems Over an Erasure Channel Abstract: The output tracking control problem is investigated in this article. First, a new tradeoff performance index is presented for single-input–multioutput (SIMO) systems. Based on the frequency-domain method, the tracking performance limitations under time delay, packet loss, and channel noise effects are derived. We use a bivariate stochastic process to model the packet loss, and assume that channel ...

Title: A Developmental Cognitive Architecture for Trust and Theory of Mind in Humanoid Robots Abstract: As artificial systems are starting to be widely deployed in real-world settings, it becomes critical to provide them with the ability to discriminate between different informants and to learn from reliable sources. Moreover, equipping an artificial agent to infer beliefs may improve the collaboration between humans and machines in several ways. In this article, we propose a hybrid cognitive archit...

Title: Absence of Pilot Monitoring Affects Scanning Behavior of Pilot Flying: Implications for the Design of Single-Pilot Cockpits Abstract: Objective This study examines whether the pilot flying's (PF) scanning behavior is affected by the absence of the pilot monitoring (PM) and aims at deriving implications for the design of single-pilot cockpits for commercial aviation. Background Due to technological progress, a crew reduction from two-crew to single-pilot operations (SPO) might be feasible. This requires a redesign of the cockpit to support the pilot adequately, especially during high workload phases such as approach and landing. In these phases, the continuous scanning of flight parameters is of particular importance. Method Experienced pilots flew various approach and landing scenarios with or without the support of the PM in a fixed-base Airbus A320 simulator. A within-subject design was used and eye-tracking data were collected to analyze scanning behavior. Results The results confirm that the absence of the PM affects the PF's scanning behavior. Participants spent significantly more time scanning secondary instruments at the expense of primary instruments when flying alone. Moreover, the frequency of transitions between the cockpit instruments and the external view increased while mean dwell durations on the external view decreased. Conclusion The findings suggest that the PM supports the PF to achieve efficient scanning behavior. Information should be presented differently in commercial SPO to compensate for the PM's absence and to avoid visual overload. Application This research will help inform the design of commercial SPO flight decks providing adequate support for the pilot particularly in terms of efficient scanning behavior.

Title: Critic Learning-Based Control for Robotic Manipulators With Prescribed Constraints Abstract: In this article, the optimal control problem for robotic manipulators (RMs) with prescribed constraints is addressed. Considering the environmental conditions and requirements of practical applications, prescribed constraints are imposed on the system states to guarantee the control performance and normal operation of the robotic system. Accordingly, an error transformation function is adopted to ...

Title: H ∞ Control for Observer-Based Non-Negative Edge Consensus of Discrete-Time Networked Systems Abstract: This article is concerned with observer-based non-negative edge-consensus (OBNNEC) problems of networked discrete-time systems with or without actuator saturation. An algorithm which only uses actual outputs of neighboring edges is proposed by means of an $H_{\infty }$ control method and modified algebraic Riccati equation (M...

Title: Learning With Selected Features Abstract: The coming big data era brings data of unprecedented size and launches an innovation of learning algorithms in statistical and machine-learning communities. The classical kernel-based regularized least-squares (RLS) algorithm is excluded in the innovation, due to its computational and storage bottlenecks. This article presents a scalable algorithm based on subsampling, called learning with ...

Title: Accommodating Multiple Tasks’ Disparities With Distributed Knowledge-Sharing Mechanism Abstract: Deep multitask learning (MTL) shares beneficial knowledge across participating tasks, alleviating the impacts of extreme learning conditions on their performances such as the data scarcity problem. In practice, participators stemming from different domain sources often have varied complexities and input sizes, for example, in the joint learning of computer vision tasks with RGB and grayscale image...

Title: Improving generalization of deep neural networks by leveraging margin distribution Abstract: Recent research has used margin theory to analyze the generalization performance for deep neural networks (DNNs). The existed results are almost based on the spectrally-normalized minimum margin. However, optimizing the minimum margin ignores a mass of information about the entire margin distribution, which is crucial to generalization performance. In this paper, we prove a generalization upper bound dominated by the statistics of the entire margin distribution. Compared with the minimum margin bounds, our bound highlights an important measure for controlling the complexity, which is the ratio of the margin standard deviation to the expected margin. We utilize a convex margin distribution loss function on the deep neural networks to validate our theoretical results by optimizing the margin ratio. Experiments and visualizations confirm the effectiveness of our approach and the correlation between generalization gap and margin ratio.