_id
sequence
label
stringlengths
5
77
text
stringlengths
78
3.43k
[ 21 ]
math.CO
Proportional 2 Choosability with a Bounded Palette. Proportional choosability is a list coloring analogue of equitable coloring. Specifically, a k assignment L for a graph G specifies a list L(v) of k available colors to each v in V(G) . An L coloring assigns a color to each vertex v from its list L(v) . A proportional L coloring of G is a proper L coloring in which each color c in bigcup v in V(G) L(v) is used lfloor eta(c) k rfloor or lceil eta(c) k rceil times where eta(c) left lvert v in V(G) c in L(v) right rvert . A graph G is proportionally k choosable if a proportional L coloring of G exists whenever L is a k assignment for G . Motivated by earlier work, we initiate the study of proportional choosability with a bounded palette by studying proportional 2 choosability with a bounded palette. In particular, when ell geq 2 , a graph G is said to be proportionally (2, ell) choosable if a proportional L coloring of G exists whenever L is a 2 assignment for G satisfying bigcup v in V(G) L(v) leq ell . We observe that a graph is proportionally (2,2) choosable if and only if it is equitably 2 colorable. As ell gets larger, the set of proportionally (2, ell) choosable graphs gets smaller. We show that whenever ell geq 5 a graph is proportionally (2, ell) choosable if and only if it is proportionally 2 choosable. We also completely characterize the connected proportionally (2, ell) choosable graphs when ell 3,4 .
[ 65, 15 ]
cs.IT cs.NI math.IT
Meta Distribution of Downlink tt SIR in a Poisson Cluster Process based HetNet Model. The performance analysis of heterogeneous cellular networks (HetNets), that relied mostly on the homogeneous Poisson point process (PPP) for the spatial distribution of the users and base stations (BSs), has seen a major transition with the emergence of the Poisson cluster process (PCP) based models. With the combination of PPP and PCP, it is possible to construct a general HetNet model which can capture formation of hotspots and spatial coupling between the users and the BSs. While the downlink coverage analysis of this model in terms of the distribution of the received downlink signal to interference ratio ( tt SIR ) is well understood by now, more fine grained analysis in terms of the meta distribution of tt SIR is an open problem. In this letter, we solve this problem by deriving the meta distribution of the downlink tt SIR assuming that the typical user connects to the BS providing the maximum received power.
[ 16, 51, 92 ]
q-bio.BM physics.bio-ph q-bio.QM
ThreshKnot Thresholded ProbKnot for Improved RNA Secondary Structure Prediction. RNA structure prediction is a challenging problem, especially with pseudoknots. Recently, there has been a shift from the classical minimum free energy based methods (MFE) to partition function based ones that assemble structures using base pairing probabilities. Two examples of the latter group are the popular maximum expected accuracy (MEA) method and the ProbKnot method. ProbKnot is a fast heuristic that pairs nucleotides that are reciprocally most probable pairing partners, and unlike MEA, can also predict structures with pseudoknots. However, ProbKnot's full potential has been largely overlooked. In particular, when introduced, it did not have an MEA like hyperparameter that can balance between positive predictive value (PPV) and sensitivity. We show that a simple thresholded version of ProbKnot, which we call ThreshKnot, leads to more accurate overall predictions by filtering out unlikely pairs whose probabilities fall under a given threshold. We also show that on three widely used folding engines (RNAstructure, Vienna RNAfold, and CONTRAfold), ThreshKnot always outperforms the much more involved MEA algorithm in (1) its higher structure prediction accuracy, (2) its capability to predict pseudoknots, and (3) its faster runtime and easier implementation. This suggests that ThreshKnot should replace MEA as the default partition function based structure prediction algorithm. ThreshKnot is already available in the widely used RNAstructure software package version 6.2 (released November 27, 2019) https rna.urmc.rochester.edu RNAstructure.html
[ 9, 12, 46 ]
math.CA math.AG math.AP
Euler and Laplace integral representations of GKZ hypergeometric functions. We introduce an interpolation between Euler integral and Laplace integral Euler Laplace integral. We establish a combinatorial method of constructing a basis of the rapid decay homology group associated to Euler Laplace integral with a nice intersection property. This construction yields a remarkable expansion formula of cohomology intersection numbers in terms of GKZ hypergeometric series. As an application, we obtain closed formulas of the quadratic relations of Aomoto Gelfand hypergeometric functions and their confluent analogue in terms of bipartite graphs.
[ 36 ]
physics.chem-ph
Energy Based Molecular Orbital Localization in a Specific Spatial Region. We present a novel energy based localization procedure able to localize molecular orbitals into specific spatial regions. The method is applied to several cases including both conjugated and non conjugated systems. The obtained localized molecular orbitals are used in a multiscale framework based on the multilevel Hartree Fock approach. An almost perfect agreement with reference values is achieved for both ground state properties, such as dipole moments, and local excitation energies calculated at the coupled cluster level. The proposed approach is useful to extend the application range of high level electron correlation methods. In fact, the reduced number of molecular orbitals can lead to a large reduction in the computational cost of correlated calculations.
[ 39 ]
cs.CY
Embedding Culture and Grit in the Technology Acceptance Model (TAM) for Higher Education. The implementors of learning technologies within education environments often follow strategies that assume the educational environment within which they are being introduced is culturally neutral. A comprehensive literature review including 150 papers on educational technology challenges was undertaken. The purpose of this review is explore different contextual challenges to the adoption of educational technology in the higher education sector. The cultural factors that define the key stakeholders (e.g., teachers, lectures, students and support staff) are often ignored when the implementation processes are undertaken. Furthermore, it is often assumed that the personnel responsible for the implementation are also culturally neutral and do not possess any attributes unique to their culture. It has been shown that cultural factors may significantly influence the implementation of learning technologies and to design strategies that fail to consider factors may limit their efficiency and effectiveness. The challenges are interrelated and based on the findings, this review proposes a conceptual framework by integrating culture and grit into the Technology Acceptance Model(TAM) for implementing educational technology in higher education. The framework will be useful to guide both practice and research.
[ 57 ]
cs.PL
ICurry. FlatCurry is a well established intermediate representation of Curry programs used in compilers that translate Curry code into Prolog and Haskell code. Some FlatCurry constructs have no direct translation into imperative code. These constructs must be each handled differently when translating Curry code into C, C and Python code. We introduce a new representation of Curry programs, called ICurry, and derive a translation from all FlatCurry constructs into ICurry. We present the syntax of ICurry and the translation from FlatCurry to ICurry. We present a model of functional logic computations as graph rewriting, show how this model can be implemented in a low level imperative language, and describe the translation from ICurry to this model.
[ 22 ]
cs.SE
The SmartSHARK Ecosystem for Software Repository Mining. Software repository mining is the foundation for many empirical software engineering studies. The collection and analysis of detailed data can be challenging, especially if data shall be shared to enable replicable research and open science practices. SmartSHARK is an ecosystem that supports replicable and reproducible research based on software repository mining.
[ 26 ]
math.RT
GKM Theory for Torus Actions on Cyclic Quiver Grassmannians. We define and investigate algebraic torus actions on quiver Grassmannians for nilpotent representations of the equioriented cycle. Examples of such varieties are type tt A flag varieties, their linear degenerations, finite dimensional approximations of the GL n affine flag variety and affine Grassmannian. We show that these quiver Grassmannians, equipped with our torus action, are GKM varieties and that their moment graph admits a combinatorial description in terms of coefficients quiver of the underlying quiver representations. By adapting to our setting results by Gonzales, we are able to prove that moment graph techniques can be applied to construct module bases for the equivariant cohomology of the above quiver Grassmannians.
[ 39 ]
cs.CY
Closing the AI Accountability Gap Defining an End to End Framework for Internal Algorithmic Auditing. Rising concern for the societal implications of artificial intelligence systems has inspired a wave of academic and journalistic literature in which deployed systems are audited for harm by investigators from outside the organizations deploying the algorithms. However, it remains challenging for practitioners to identify the harmful repercussions of their own systems prior to deployment, and, once deployed, emergent issues can become difficult or impossible to trace back to their source. In this paper, we introduce a framework for algorithmic auditing that supports artificial intelligence system development end to end, to be applied throughout the internal organization development lifecycle. Each stage of the audit yields a set of documents that together form an overall audit report, drawing on an organization's values or principles to assess the fit of decisions made throughout the process. The proposed auditing framework is intended to contribute to closing the accountability gap in the development and deployment of large scale artificial intelligence systems by embedding a robust process to ensure audit integrity.
[ 25 ]
math.PR
Implementation of Strong Numerical Methods of Orders 0.5, 1.0, 1.5, 2.0, 2.5, and 3.0 for Ito SDEs with Non Commutative Noise Based on the Unified Taylor Ito and Taylor Stratonovich Expansions and Multiple Fourier Legendre Series. The article is devoted to the implementation of strong numerical methods with convergence orders 0.5, 1.0, 1.5, 2.0, 2.5, and 3.0 for Ito stochastic differential equations with multidimensional non commutative noise based on the unified Taylor Ito and Taylor Stratonovich expansions and multiple Fourier Legendre series. Algorithms for the implementation of these methods are constructed and a package of programs in the Python programming language is presented. An important part of this software package, concerning the mean square approximation of iterated Ito and Stratonovich stochastic integrals of multiplicities 1 to 6 with respect to components of the multidimensional Wiener process is based on the method of generalized multiple Fourier series. More precisely, we used the multiple Fourier Legendre series converging in the sense of norm in Hilbert space L 2( t, T k) (k 1, ldots,6) for the mean square approximation of iterated Ito and Stratonovich stochastic integrals.
[ 65, 1 ]
cs.LG cs.NI
Pareto Deterministic Policy Gradients and Its Application in 5G Massive MIMO Networks. In this paper, we consider jointly optimizing cell load balance and network throughput via a reinforcement learning (RL) approach, where inter cell handover (i.e., user association assignment) and massive MIMO antenna tilting are configured as the RL policy to learn. Our rationale behind using RL is to circumvent the challenges of analytically modeling user mobility and network dynamics. To accomplish this joint optimization, we integrate vector rewards into the RL value network and conduct RL action via a separate policy network. We name this method as Pareto deterministic policy gradients (PDPG). It is an actor critic, model free and deterministic policy algorithm which can handle the coupling objectives with the following two merits 1) It solves the optimization via leveraging the degree of freedom of vector reward as opposed to choosing handcrafted scalar reward 2) Cross validation over multiple policies can be significantly reduced. Accordingly, the RL enabled network behaves in a self organized way It learns out the underlying user mobility through measurement history to proactively operate handover and antenna tilt without environment assumptions. Our numerical evaluation demonstrates that the introduced RL method outperforms scalar reward based approaches. Meanwhile, to be self contained, an ideal static optimization based brute force search solver is included as a benchmark. The comparison shows that the RL approach performs as well as this ideal strategy, though the former one is constrained with limited environment observations and lower action frequency, whereas the latter ones have full access to the user mobility. The convergence of our introduced approach is also tested under different user mobility environment based on our measurement data from a real scenario.
[ 21 ]
math.CO
Wiener Index and Remoteness in Triangulations and Quadrangulations. Let G be a a connected graph. The Wiener index of a connected graph is the sum of the distances between all unordered pairs of vertices. We provide asymptotic formulae for the maximum Wiener index of simple triangulations and quadrangulations with given connectivity, as the order increases, and make conjectures for the extremal triangulations and quadrangulations based on computational evidence. If overline sigma (v) denotes the arithmetic mean of the distances from v to all other vertices of G , then the remoteness of G is defined as the largest value of overline sigma (v) over all vertices v of G . We give sharp upper bounds on the remoteness of simple triangulations and quadrangulations of given order and connectivity.
[ 40 ]
cs.IR
Semantic Product Search for Matching Structured Product Catalogs in E Commerce. Retrieving all semantically relevant products from the product catalog is an important problem in E commerce. Compared to web documents, product catalogs are more structured and sparse due to multi instance fields that encode heterogeneous aspects of products (e.g. brand name and product dimensions). In this paper, we propose a new semantic product search algorithm that learns to represent and aggregate multi instance fields into a document representation using state of the art transformers as encoders. Our experiments investigate two aspects of the proposed approach (1) effectiveness of field representations and structured matching (2) effectiveness of adding lexical features to semantic search. After training our models using user click logs from a well known E commerce platform, we show that our results provide useful insights for improving product search. Lastly, we present a detailed error analysis to show which types of queries benefited the most by fielded representations and structured matching.
[ 40, 1 ]
cs.IR cs.LG
A Survey on Deep Learning Based Point Of Interest (POI) Recommendations. Location based Social Networks (LBSNs) enable users to socialize with friends and acquaintances by sharing their check ins, opinions, photos, and reviews. Huge volume of data generated from LBSNs opens up a new avenue of research that gives birth to a new sub field of recommendation systems, known as Point of Interest (POI) recommendation. A POI recommendation technique essentially exploits users' historical check ins and other multi modal information such as POI attributes and friendship network, to recommend the next set of POIs suitable for a user. A plethora of earlier works focused on traditional machine learning techniques by using hand crafted features from the dataset. With the recent surge of deep learning research, we have witnessed a large variety of POI recommendation works utilizing different deep learning paradigms. These techniques largely vary in problem formulations, proposed techniques, used datasets, and features, etc. To the best of our knowledge, this work is the first comprehensive survey of all major deep learning based POI recommendation works. Our work categorizes and critically analyzes the recent POI recommendation works based on different deep learning paradigms and other relevant features. This review can be considered a cookbook for researchers or practitioners working in the area of POI recommendation.
[ 41, 14 ]
cs.RO cs.HC
Gait Graph Optimization Generate Variable Gaits from One Base Gait for Lower limb Rehabilitation Exoskeleton Robots. The most concentrated application of lower limb rehabilitation exoskeleton (LLE) robot is that it can help paraplegics "re walk". However, "walking" in daily life is more than just walking on flat ground with fixed gait. This paper focuses on variable gaits generation for LLE robot to adapt complex walking environment. Different from traditional gaits generator for biped robot, the generated gaits for LLEs should be comfortable to patients. Inspired by the pose graph optimization algorithm in SLAM, we propose a graph based gait generation algorithm called gait graph optimization (GGO) to generate variable, functional and comfortable gaits from one base gait collected from healthy individuals to adapt the walking environment. Variants of walking problem, e.g., stride adjustment, obstacle avoidance, and stair ascent and descent, help verify the proposed approach in simulation and experimentation. We open source our implementation.
[ 1, 3 ]
cs.CL cs.LG
MKD a Multi Task Knowledge Distillation Approach for Pretrained Language Models. Pretrained language models have led to significant performance gains in many NLP tasks. However, the intensive computing resources to train such models remain an issue. Knowledge distillation alleviates this problem by learning a light weight student model. So far the distillation approaches are all task specific. In this paper, we explore knowledge distillation under the multi task learning setting. The student is jointly distilled across different tasks. It acquires more general representation capacity through multi tasking distillation and can be further fine tuned to improve the model in the target domain. Unlike other BERT distillation methods which specifically designed for Transformer based architectures, we provide a general learning framework. Our approach is model agnostic and can be easily applied on different future teacher model architectures. We evaluate our approach on a Transformer based and LSTM based student model. Compared to a strong, similarly LSTM based approach, we achieve better quality under the same computational constraints. Compared to the present state of the art, we reach comparable results with much faster inference speed.
[ 0 ]
cs.CV
Dynamic Graph Learning Instance aware Connectivity for Neural Networks. One practice of employing deep neural networks is to apply the same architecture to all the input instances. However, a fixed architecture may not be representative enough for data with high diversity. To promote the model capacity, existing approaches usually employ larger convolutional kernels or deeper network structure, which may increase the computational cost. In this paper, we address this issue by raising the Dynamic Graph Network (DG Net). The network learns the instance aware connectivity, which creates different forward paths for different instances. Specifically, the network is initialized as a complete directed acyclic graph, where the nodes represent convolutional blocks and the edges represent the connection paths. We generate edge weights by a learnable module textit router and select the edges whose weights are larger than a threshold, to adjust the connectivity of the neural network structure. Instead of using the same path of the network, DG Net aggregates features dynamically in each node, which allows the network to have more representation ability. To facilitate the training, we represent the network connectivity of each sample in an adjacency matrix. The matrix is updated to aggregate features in the forward pass, cached in the memory, and used for gradient computing in the backward pass. We verify the effectiveness of our method with several static architectures, including MobileNetV2, ResNet, ResNeXt, and RegNet. Extensive experiments are performed on ImageNet classification and COCO object detection, which shows the effectiveness and generalization ability of our approach.
[ 19, 1, 3 ]
cs.LG cs.CL cs.CR
Differentially Private Language Models Benefit from Public Pre training. Language modeling is a keystone task in natural language processing. When training a language model on sensitive information, differential privacy (DP) allows us to quantify the degree to which our private data is protected. However, training algorithms which enforce differential privacy often lead to degradation in model quality. We study the feasibility of learning a language model which is simultaneously high quality and privacy preserving by tuning a public base model on a private corpus. We find that DP fine tuning boosts the performance of language models in the private domain, making the training of such models possible.
[ 21 ]
math.CO
Promotion Sorting. Sch " u tzenberger's promotion operator is an extensively studied bijection that permutes the linear extensions of a finite poset. We introduce a natural extension partial of this operator that acts on all labelings of a poset. We prove several properties of partial in particular, we show that for every labeling L of an n element poset P , the labeling partial n 1 (L) is a linear extension of P . Thus, we can view the dynamical system defined by partial as a sorting procedure that sorts labelings into linear extensions. For all 0 leq k leq n 1 , we characterize the n element posets P that admit labelings that require at least n k 1 iterations of partial in order to become linear extensions. The case in which k 0 concerns labelings that require the maximum possible number of iterations in order to be sorted we call these labelings tangled. We explicitly enumerate tangled labelings for a large class of posets that we call inflated rooted forest posets. For an arbitrary finite poset, we show how to enumerate the sortable labelings, which are the labelings L such that partial(L) is a linear extension.
[ 40, 3, 63 ]
cs.CL cs.DL cs.IR
Causal Knowledge Extraction from Scholarly Papers in Social Sciences. The scale and scope of scholarly articles today are overwhelming human researchers who seek to timely digest and synthesize knowledge. In this paper, we seek to develop natural language processing (NLP) models to accelerate the speed of extraction of relationships from scholarly papers in social sciences, identify hypotheses from these papers, and extract the cause and effect entities. Specifically, we develop models to 1) classify sentences in scholarly documents in business and management as hypotheses (hypothesis classification), 2) classify these hypotheses as causal relationships or not (causality classification), and, if they are causal, 3) extract the cause and effect entities from these hypotheses (entity extraction). We have achieved high performance for all the three tasks using different modeling techniques. Our approach may be generalizable to scholarly documents in a wide range of social sciences, as well as other types of textual materials.
[ 11 ]
physics.app-ph
Steering Flexural Waves by Amplitude Shift Elastic Metasurfaces. As 2D materials with subwavelength structures, elastic metasurfaces show remarkable abilities to manipulate elastic waves at will through artificial boundary conditions. However, the application prospects of current metasurfaces may be restricted by their phase only modulating boundaries. Herein, we present the next generation of elastic metasurfaces by additionally incorporating amplitude shift modulation. A general theory for target wave fields steered by metasurfaces is proposed by modifying the Huygens Fresnel principle. As examples, two amplitude shift metasurfaces concerning flexural waves in thin plates are carried out one is to transform a cylindrical wave into a Gaussian beam by elaborating both amplitude and phase shifts, and the other one is to focus the incidence by amplitude modulations only. These examples coincide well over theoretical calculations, numerical simulations and experimental tests. This work may underlie the design of metasurfaces with complete control over guided elastic waves, and may extend to more sophisticated applications, such as analog signal processing and holographic imaging.
[ 21 ]
math.CO
Double Affine Bruhat Order. We classify cocovers and covers of a given element of the double affine Weyl semigroup W with respect to the Bruhat order, specifically when W is associated to a finite root system that is irreducible and simply laced. We show two approaches one extending the work of Lam and Shimozono, and its strengthening by Milicevic, where cocovers are characterized in the affine case using the quantum Bruhat graph of the finite Weyl group, and another, which takes a more geometrical approach by using the length difference set defined by Muthiah and Orr.
[ 23 ]
math.FA
Functional L "owner Ellipsoids. We extend the notion of the smallest volume ellipsoid containing a convex body in mathbb R d to the setting of logarithmically concave functions. We consider a vast class of logarithmically concave functions whose superlevel sets are concentric ellipsoids. For a fixed function from this class, we consider the set of all its "affine" positions. For any log concave function f on mathbb R d , we consider functions belonging to this set of "affine" positions, and find the one with the smallest integral under the condition that it is pointwise greater than or equal to f. We study the properties of existence and uniqueness of the solution to this problem. For any s in 0, infty), we consider the construction dual to the recently defined John s function cite ivanov2020functional . We prove that such a construction determines a unique function and call it the emph L "owner s function of f. We study the L "owner s functions as s tends to zero and to infinity. Finally, extending the notion of the outer volume ratio, we define the outer integral ratio of a log concave function and give an asymptotically tight bound on it. end abstract
[ 1, 3 ]
cs.LG cs.CL
TCNN Triple Convolutional Neural Network Models for Retrieval based Question Answering System in E commerce. Automatic question answering (QA) systems have boomed during last few years, and commonly used techniques can be roughly categorized into Information Retrieval (IR) based and generation based. A key solution to the IR based models is to retrieve the most similar knowledge entries of a given query from a QA knowledge base, and then rerank those knowledge entries with semantic matching models. In this paper, we aim to improve an IR based e commerce QA system AliMe with proposed text matching models, including a basic Triple Convolutional Neural Network (TCNN) model and two Attention based TCNN (ATCNN) models. Experimental results show their effect.
[ 24 ]
math.MG
Volumetric bounds for intersections of congruent balls. We investigate the intersections of balls of radius r , called r ball bodies, in Euclidean d space. An r lense (resp., r spindle) is the intersection of two balls of radius r (resp., balls of radius r containing a given pair of points). We prove that among r ball bodies of given volume, the r lense (resp., r spindle) has the smallest inradius (resp., largest circumradius). In general, we upper (resp., lower) bound the intrinsic volumes of r ball bodies of given inradius (resp., circumradius). This complements and extends some earlier results on volumetric estimates for r ball bodies.
[ 21 ]
math.CO
On distinct consecutive differences. We show that if A a 1 a 2 ldots a k is a set of real numbers such that the differences of the consecutive elements are distinct, then for and finite B subset mathbb R , A B gg A 1 2 B . The bound is tight up to the constant.
[ 41, 2, 1 ]
cs.RO cs.AI cs.LG
Motion Planner Augmented Reinforcement Learning for Robot Manipulation in Obstructed Environments. Deep reinforcement learning (RL) agents are able to learn contact rich manipulation tasks by maximizing a reward signal, but require large amounts of experience, especially in environments with many obstacles that complicate exploration. In contrast, motion planners use explicit models of the agent and environment to plan collision free paths to faraway goals, but suffer from inaccurate models in tasks that require contacts with the environment. To combine the benefits of both approaches, we propose motion planner augmented RL (MoPA RL) which augments the action space of an RL agent with the long horizon planning capabilities of motion planners. Based on the magnitude of the action, our approach smoothly transitions between directly executing the action and invoking a motion planner. We evaluate our approach on various simulated manipulation tasks and compare it to alternative action spaces in terms of learning efficiency and safety. The experiments demonstrate that MoPA RL increases learning efficiency, leads to a faster exploration, and results in safer policies that avoid collisions with the environment. Videos and code are available at https clvrai.com mopa rl .
[ 94 ]
physics.ed-ph
Teaching and learning special relativity theory in secondary and lower undergraduate education A literature review. This review presents an overview and analysis of the body of research on special relativity theory (SRT) education at the secondary and lower undergraduate level. There is currently a growing international interest in implementing SRT in pre university education as an introduction to modern physics. For this reason, insights into learning opportunities and challenges in SRT education are needed. The field of research in SRT education is still at an early stage, especially at the level of secondary education, and there is a shortage of empirical evaluation of learning outcomes. In order to guide future research directions, there is a need for an overview and synthesis of the results reported so far. We have selected 40 articles and categorized them according to reported learning difficulties, teaching approaches, and research tools. Analysis shows that students at all educational levels experience learning difficulties with the use of frames of reference, the postulates of SRT, and relativistic effects. In the reported teaching sequences, instructional materials, and learning activities, these difficulties are approached from different angles. Some teaching approaches focus on thought experiments to express conceptual features of SRT, while others use virtual environments to provide realistic visualization of relativistic effects. From the reported teaching approaches, three learning objectives can be identified to foster conceptual understanding, to foster understanding of the history and philosophy of science, and to gain motivation and confidence toward SRT and physics in general. In order to quantitatively compare learning outcomes of different teaching strategies, a more thorough evaluation of assessment tools is required.
[ 46 ]
math.AG
Description of generalized isomonodromic deformations of rank two linear differential equations using apparent singularities. In this paper, we consider the generalized isomonodromic deformations of rank two irregular connections on the Riemann sphere. We introduce Darboux coordinates on the parameter space of a family of rank two irregular connections by apparent singularities. By the Darboux coordinates, we describe the generalized isomonodromic deformations as Hamiltonian systems.
[ 43, 19, 17 ]
cs.CR cs.DB cs.DC
Enjoy the Untrusted Cloud A Secure, Scalable and Efficient SQL like Query Framework for Outsourcing Data. While the security of the cloud remains a concern, a common practice is to encrypt data before outsourcing them for utilization. One key challenging issue is how to efficiently perform queries over the ciphertext. Conventional crypto based solutions, e.g. partially fully homomorphic encryption and searchable encryption, suffer from low performance, poor expressiveness and weak compatibility. An alternative method that utilizes hardware assisted trusted execution environment, i.e., Intel SGX, has emerged recently. On one hand, such work lacks of supporting scalable access control over multiple data users. On the other hand, existing solutions are subjected to the key revocation problem and knowledge extractor vulnerability. In this work, we leverage the newly hardware assisted methodology and propose a secure, scalable and efficient SQL like query framework named QShield. Building upon Intel SGX, QShield can guarantee the confidentiality and integrity of sensitive data when being processed on an untrusted cloud platform. Moreover, we present a novel lightweight secret sharing method to enable multi user access control in QShield, while tackling the key revocation problem. Furthermore, with an additional trust proof mechanism, QShield guarantees the correctness of queries and significantly alleviates the possibility to build a knowledge extractor. We implemented a prototype for QShield and show that QShield incurs minimum performance cost.
[ 21 ]
math.CO
Counting Labeled Threshold Graphs with Eulerian Numbers. A threshold graph is any graph which can be constructed from the empty graph by repeatedly adding a new vertex that is either adjacent to every vertex or to no vertices. The Eulerian number genfrac langle rangle 0pt n k counts the number of permutations of size n with exactly k ascents. Implicitly Beissinger and Peled proved that the number of labeled threshold graphs on n ge 2 vertices is sum k 1 n 1 (n k) genfrac langle rangle 0pt n 1 k 1 2 k. Their proof used generating functions. We give a direct combinatorial proof of this result.
[ 3 ]
cs.CL
Generating Synthetic Data for Task Oriented Semantic Parsing with Hierarchical Representations. Modern conversational AI systems support natural language understanding for a wide variety of capabilities. While a majority of these tasks can be accomplished using a simple and flat representation of intents and slots, more sophisticated capabilities require complex hierarchical representations supported by semantic parsing. State of the art semantic parsers are trained using supervised learning with data labeled according to a hierarchical schema which might be costly to obtain or not readily available for a new domain. In this work, we explore the possibility of generating synthetic data for neural semantic parsing using a pretrained denoising sequence to sequence model (i.e., BART). Specifically, we first extract masked templates from the existing labeled utterances, and then fine tune BART to generate synthetic utterances conditioning on the extracted templates. Finally, we use an auxiliary parser (AP) to filter the generated utterances. The AP guarantees the quality of the generated data. We show the potential of our approach when evaluating on the Facebook TOP dataset for navigation domain.
[ 0, 83 ]
cs.CV cs.MM
Learning to Sample the Most Useful Training Patches from Images. Some image restoration tasks like demosaicing require difficult training samples to learn effective models. Existing methods attempt to address this data training problem by manually collecting a new training dataset that contains adequate hard samples, however, there are still hard and simple areas even within one single image. In this paper, we present a data driven approach called PatchNet that learns to select the most useful patches from an image to construct a new training set instead of manual or random selection. We show that our simple idea automatically selects informative samples out from a large scale dataset, leading to a surprising 2.35dB generalisation gain in terms of PSNR. In addition to its remarkable effectiveness, PatchNet is also resource friendly as it is applied only during training and therefore does not require any additional computational cost during inference.
[ 18, 29 ]
math.GT math.GR
Spectral gap of scl in graphs of groups and 3 manifolds. The stable commutator length scl G(g) of an element g in a group G is an invariant for group elements sensitive to the geometry and dynamics of G. For any group G acting on a tree, we prove a sharp bound scl G(g) 1 2 for any g acting without fixed points, provided that the stabilizer of each edge is relatively torsion free in its vertex stabilizers. The sharp gap becomes 1 2 1 n if the edge stabilizers are n relatively torsion free in vertex stabilizers. We also compute scl G for elements acting with a fixed point. This implies many such groups have a spectral gap, that is, there is a constant C 0 such that either scl G(g) C or scl G(g) 0. New examples include the fundamental group of any 3 manifold using the JSJ decomposition, though the gap must depend on the manifold. We also compute the exact gap of graph products. We prove these statements by characterizing maps of surfaces to a suitable K(G,1). In many cases, we also find families of quasimorphisms that realize these gaps. In particular, we construct quasimorphisms realizing the 1 2 gap for free groups explicitly induced by actions on the circle.
[ 0 ]
cs.CV
Adversarial Network with Multiple Classifiers for Open Set Domain Adaptation. Domain adaptation aims to transfer knowledge from a domain with adequate labeled samples to a domain with scarce labeled samples. Prior research has introduced various open set domain adaptation settings in the literature to extend the applications of domain adaptation methods in real world scenarios. This paper focuses on the type of open set domain adaptation setting where the target domain has both private ('unknown classes') label space and the shared ('known classes') label space. However, the source domain only has the 'known classes' label space. Prevalent distribution matching domain adaptation methods are inadequate in such a setting that demands adaptation from a smaller source domain to a larger and diverse target domain with more classes. For addressing this specific open set domain adaptation setting, prior research introduces a domain adversarial model that uses a fixed threshold for distinguishing known from unknown target samples and lacks at handling negative transfers. We extend their adversarial model and propose a novel adversarial domain adaptation model with multiple auxiliary classifiers. The proposed multi classifier structure introduces a weighting module that evaluates distinctive domain characteristics for assigning the target samples with weights which are more representative to whether they are likely to belong to the known and unknown classes to encourage positive transfers during adversarial training and simultaneously reduces the domain gap between the shared classes of the source and target domains. A thorough experimental investigation shows that our proposed method outperforms existing domain adaptation methods on a number of domain adaptation datasets.
[ 74, 27, 77, 23 ]
math.CT math.FA math.LO math.OA
Hilbert spaces and C ast algebras are not finitely concrete. We show that no faithful functor from the category of Hilbert spaces with linear isometries into the category of sets preserves directed colimits. Thus Hilbert spaces cannot form an abstract elementary class, even up to change of language. We deduce an analogous result for the category of commutative unital C ast algebras with ast homomorphisms. This implies, in particular, that this category is not axiomatizable by a first order theory, a strengthening of a conjecture of Bankston.
[ 70 ]
math.HO
An encounter in the realm of Structural Stability between a qualitative theory for geometric shapes and one for the integral foliations of differential equations. This evocative essay focuses on some landmarks that led the author to the study of principal curvature configurations on surfaces in mathbb R 3 , their structural stability and generic properties. The starting point was an encounter with the book of D. Struik and the reading of the references to the works of Euler, Monge and Darboux found there. The concatenation of these references with the work of Peixoto, 1962, on differential equations on surfaces, was a crucial second step. The circumstances of the convergence toward the theorems of Guti 'errez and Sotomayor, 1982 1983, are recounted here. The above 1982 1983 theorems are pointed out as the first encounter between the line of thought disclosed from the works of Monge, 1796, Dupin, 1815, and Darboux, 1896, with that transpiring from the achievements of Poincar 'e, 1881, Andronov Pontrjagin, 1937, and Peixoto, 1962. Some mathematical developments sprouting from the 1982 1983 works are mentioned on the final section of this essay.
[ 9 ]
math.AP
Estimate on the dimension of the singular set of the supercritical surface quasigeostrophic equation. We consider the SQG equation with dissipation given by a fractional Laplacian of order alpha frac 1 2 . We introduce a notion of suitable weak solution, which exists for every L 2 initial datum, and we prove that for such solution the singular set is contained in a compact set in spacetime of Hausdorff dimension at most frac 1 2 alpha left( frac 1 alpha alpha (1 2 alpha) 2 right) .
[ 56, 77 ]
math.LO cs.LO
A characterisation of ordered abstract probabilities. In computer science, especially when dealing with quantum computing or other non standard models of computation, basic notions in probability theory like "a predicate" vary wildly. There seems to be one constant the only useful example of an algebra of probabilities is the real unit interval. In this paper we try to explain this phenomenon. We will show that the structure of the real unit interval naturally arises from a few reasonable assumptions. We do this by studying effect monoids, an abstraction of the algebraic structure of the real unit interval it has an addition x y which is only defined when x y leq 1 and an involution x mapsto 1 x which make it an effect algebra, in combination with an associative (possibly non commutative) multiplication. Examples include the unit intervals of ordered rings and Boolean algebras. We present a structure theory for effect monoids that are omega complete, i.e. where every increasing sequence has a supremum. We show that any omega complete effect monoid embeds into the direct sum of a Boolean algebra and the unit interval of a commutative unital C algebra. This gives us from first principles a dichotomy between sharp logic, represented by the Boolean algebra part of the effect monoid, and probabilistic logic, represented by the commutative C algebra. Some consequences of this characterisation are that the multiplication must always be commutative, and that the unique omega complete effect monoid without zero divisors and more than 2 elements must be the real unit interval. Our results give an algebraic characterisation and motivation for why any physical or logical theory would represent probabilities by real numbers.
[ 1, 22 ]
cs.LG cs.SE
Understanding the Nature of System Related Issues in Machine Learning Frameworks An Exploratory Study. Modern systems are built using development frameworks. These frameworks have a major impact on how the resulting system executes, how configurations are managed, how it is tested, and how and where it is deployed. Machine learning (ML) frameworks and the systems developed using them differ greatly from traditional frameworks. Naturally, the issues that manifest in such frameworks may differ as well as may the behavior of developers addressing those issues. We are interested in characterizing the system related issues issues impacting performance, memory and resource usage, and other quality attributes that emerge in ML frameworks, and how they differ from those in traditional frameworks. We have conducted a moderate scale exploratory study analyzing real world system related issues from 10 popular machine learning frameworks. Our findings offer implications for the development of machine learning systems, including differences in the frequency of occurrence of certain issue types, observations regarding the impact of debate and time on issue correction, and differences in the specialization of developers. We hope that this exploratory study will enable developers to improve their expectations, plan for risk, and allocate resources accordingly when making use of the tools provided by these frameworks to develop ML based systems.
[ 19, 39 ]
cs.CR cs.CY
Security in biometric systems. The objective of biometric systems is to provide an identification mechanism. This identification mechanism can be used to fulfil several objectives. The most common, related to providing security to a resource, is usually authentication or detection of authorized personnel and detection of unauthorized personnel. From the technical point of view, these two objectives can be included in a single point since most functionalities are achieved by making searches of people previously identified in the database of the system in question. In the first case access is given to people entered in the database and in the second case access is given to people who are not entered in the database. Although these are the two most common attacks there are also others that we will discuss in this chapter. The structure of the chapter is as follows. The first part of the chapter gives an overview of the basic types of attacks and describes the usual protection measures (Sections 1, 2 and 3). The second part of the chapter describes several attacks that can be made on systems based on fingerprinting, face recognition, and iris recognition (Sections 4 and 5). Once the attack methodologies have been described, some specific protection measures are also discussed (Sections 4 and 5). Finally, side channel attacks and their usefulness in combination with other possible attacks are described (Section 6).
[ 56 ]
cs.LO
A formalisation of the SPARC TSO memory model for multi core machine code. SPARC processors have many applications in mission critical industries such as aviation and space engineering. Hence, it is important to provide formal frameworks that facilitate the verification of hardware and software that run on or interface with these processors. This paper presents the first mechanised SPARC Total Store Ordering (TSO) memory model which operates on top of an abstract model of the SPARC Instruction Set Architecture (ISA) for multi core processors. Both models are specified in the theorem prover Isabelle HOL. We formalise two TSO memory models one is an adaptation of the axiomatic SPARC TSO model, the other is a novel operational TSO model which is suitable for verifying execution results. We prove that the operational model is sound and complete with respect to the axiomatic model. Finally, we give verification examples with two case studies drawn from the SPARCv9 manual.
[ 29, 46 ]
math.AG math.GR
Actions of Cremona groups on CAT(0) cube complexes. For each d we construct CAT(0) cube complexes on which Cremona groups rank d act by isometries. From these actions we deduce new and old group theoretical and dynamical results about Cremona groups. In particular, we study the dynamical behaviour of the irreducible components of exceptional loci, we prove regularization theorems, we find new constraints on the degree growth for non regularizable birational transformations, and we show that the centralizer of certain birational transformations is small.
[ 21, 38 ]
math.NT math.CO
Breaking the logarithmic barrier in Roth's theorem on arithmetic progressions. We show that if A subset 1, ldots,N contains no non trivial three term arithmetic progressions then lvert A rvert ll N ( log N) 1 c for some absolute constant c 0 . In particular, this proves the first non trivial case of a conjecture of Erd H o s on arithmetic progressions.
[ 17, 1 ]
cs.DC cs.LG
Automatic Model Parallelism for Deep Neural Networks with Compiler and Hardware Support. The deep neural networks (DNNs) have been enormously successful in tasks that were hitherto in the human only realm such as image recognition, and language translation. Owing to their success the DNNs are being explored for use in ever more sophisticated tasks. One of the ways that the DNNs are made to scale for the complex undertakings is by increasing their size deeper and wider networks can model well the additional complexity. Such large models are trained using model parallelism on multiple compute devices such as multi GPUs and multi node systems. In this paper, we develop a compiler driven approach to achieve model parallelism. We model the computation and communication costs of a dataflow graph that embodies the neural network training process and then, partition the graph using heuristics in such a manner that the communication between compute devices is minimal and we have a good load balance. The hardware scheduling assistants are proposed to assist the compiler in fine tuning the distribution of work at runtime.
[ 49 ]
math.OC
An observer for partially obstructed wood particles in industrial drying processes. In order for biomass drying processes to be efficient, it is crucial to achieve the target residual water content within a close margin, since more conservative drying would result in a waste of energy. A method for a reliable estimation of the water content is therefore of obvious importance. Ideally, such a method does not require any expensive sensors. We show reduced order models and extended Kalman filters can be combined to reliably determine the water content and temperature of wood particles based on only surface temperature measurements. The proposed observer works reliably if measurements are only available for parts of a particle face. It can therefore still be applied if particle surfaces are partially obstructed, which is a prerequisite for use in industrial processes and units, such as rotary dryers. The extended Kalman filter uses a reduced order model that is obtained by applying proper orthogonal decomposition and Galerkin projection to coupled PDEs that model heat conduction and water diffusion in anisotropic particles. In contrast to the original PDE simulation model, the reduced model and the filter based on it are suitable for real time computations and monitoring.
[ 18, 29 ]
math.GR math.GT
Automorphismes du groupe des automorphismes d'un groupe de Coxeter universel. Using the Guirardel Levitt outer space of a free product, we prove that the outer automorphism group of the outer automorphism group of the universal Coxeter group of rank n geq 5 is trivial, and that it is a cyclic group of order 2 if n 4 . In addition we prove that the outer automorphism group of the automorphism group of the universal Coxeter group of rank n geq 4 is trivial.
[ 49, 20, 25 ]
math.OC cs.NA math.NA math.PR
Tensor structured sketching for constrained least squares. Constrained least squares problems arise in many applications. Their memory and computation costs are expensive in practice involving high dimensional input data. We employ the so called "sketching" strategy to project the least squares problem onto a space of a much lower "sketching dimension" via a random sketching matrix. The key idea of sketching is to reduce the dimension of the problem as much as possible while maintaining the approximation accuracy. Tensor structure is often present in the data matrices of least squares, including linearized inverse problems and tensor decompositions. In this work, we utilize a general class of row wise tensorized sub Gaussian matrices as sketching matrices in constrained optimizations for the sketching design's compatibility with tensor structures. We provide theoretical guarantees on the sketching dimension in terms of error criterion and probability failure rate. In the context of unconstrained linear regressions, we obtain an optimal estimate for the sketching dimension. For optimization problems with general constraint sets, we show that the sketching dimension depends on a statistical complexity that characterizes the geometry of the underlying problems. Our theories are demonstrated in a few concrete examples, including unconstrained linear regression and sparse recovery problems.
[ 7 ]
physics.comp-ph
On the Spectral Behavior and Normalization of a Resonance Free and High Frequency Stable Integral Equation. The CFIE used for solving scattering and radiation problems, although a resonance free formulation, suffers from an ill conditioning that strongly depends on the frequency and discretization density, both in the low and high frequency regime, resulting in slow convergence rates for iterative solvers. This work presents a new preconditioning scheme for the CFIE that cures the low and the high frequency as well as the dense discretization breakdown. The new preconditioner for the CFIE is based on a spherical harmonics analysis and the proper regularization with Helmholtz type operators. Numerical results have been obtained to prove the effectiveness of this new formulation in real scenarios.
[ 36, 7 ]
physics.chem-ph physics.comp-ph
Isotope effects on molecular structures and electronic properties of liquid water via deep potential molecular dynamics based on SCAN functional. Feynman path integral deep potential molecular dynamics (PI DPMD) calculations have been employed to study both light (H 2 O) and heavy water (D 2 O) within the isothermal isobaric ensemble. In particular, the deep neural network is trained based on ab initio data obtained from the strongly constrained and appropriately normed (SCAN) exchange correlation functional. Because of the lighter mass of hydrogen than deuteron, the properties of light water is more influenced by nuclear quantum effect than those of heavy water. Clear isotope effects are observed and analyzed in terms of hydrogen bond structure and electronic properties of water that are closely associated with experimental observables. The molecular structures of both liquid H 2 O and D 2 O agree well with the data extracted from scattering experiments. The delicate isotope effects on radial distribution functions and angular distribution functions are well reproduced as well. Our approach demonstrates that deep neural network combined with SCAN functional based ab initio molecular dynamics provides an accurate theoretical tool for modeling water and its isotope effects.
[ 1, 2 ]
cs.LG cs.AI
Cross Domain Latent Modulation for Variational Transfer Learning. We propose a cross domain latent modulation mechanism within a variational autoencoders (VAE) framework to enable improved transfer learning. Our key idea is to procure deep representations from one data domain and use it as perturbation to the reparameterization of the latent variable in another domain. Specifically, deep representations of the source and target domains are first extracted by a unified inference model and aligned by employing gradient reversal. Second, the learned deep representations are cross modulated to the latent encoding of the alternate domain. The consistency between the reconstruction from the modulated latent encoding and the generation using deep representation samples is then enforced in order to produce inter class alignment in the latent space. We apply the proposed model to a number of transfer learning tasks including unsupervised domain adaptation and image toimage translation. Experimental results show that our model gives competitive performance.
[ 4 ]
cs.NE
Maximizing Drift is Not Optimal for Solving OneMax. It may seem very intuitive that for the maximization of the OneMax problem OM(x) sum i 1 n x i the best that an elitist unary unbiased search algorithm can do is to store a best so far solution, and to modify it with the operator that yields the best possible expected progress in function value. This assumption has been implicitly used in several empirical works. In Doerr, Doerr, Yang Optimal parameter choices via precise black box analysis, TCS, 2020 it was formally proven that this approach is indeed almost optimal. In this work we prove that drift maximization is not optimal. More precisely, we show that for most fitness levels between n 2 and 2n 3 the optimal mutation strengths are larger than the drift maximizing ones. This implies that the optimal RLS is more risk affine than the variant maximizing the step wise expected progress. We show similar results for the mutation rates of the classic (1 1) Evolutionary Algorithm (EA) and its resampling variant, the (1 1) EA 0 . As a result of independent interest we show that the optimal mutation strengths, unlike the drift maximizing ones, can be even.
[ 1, 59 ]
cs.LG cs.SI
Robust Hierarchical Graph Classification with Subgraph Attention. Graph neural networks get significant attention for graph representation and classification in machine learning community. Attention mechanism applied on the neighborhood of a node improves the performance of graph neural networks. Typically, it helps to identify a neighbor node which plays more important role to determine the label of the node under consideration. But in real world scenarios, a particular subset of nodes together, but not the individual pairs in the subset, may be important to determine the label of the graph. To address this problem, we introduce the concept of subgraph attention for graphs. On the other hand, hierarchical graph pooling has been shown to be promising in recent literature. But due to noisy hierarchical structure of real world graphs, not all the hierarchies of a graph play equal role for graph classification. Towards this end, we propose a graph classification algorithm called SubGattPool which jointly learns the subgraph attention and employs two different types of hierarchical attention mechanisms to find the important nodes in a hierarchy and the importance of individual hierarchies in a graph. Experimental evaluation with different types of graph classification algorithms shows that SubGattPool is able to improve the state of the art or remains competitive on multiple publicly available graph classification datasets. We conduct further experiments on both synthetic and real world graph datasets to justify the usefulness of different components of SubGattPool and to show its consistent performance on other downstream tasks.
[ 67 ]
physics.atom-ph
Laser Cooled Polyatomic Molecules for Improved Electron Electric Dipole Moment Searches. Doppler and Sisyphus cooling of 174 YbOH are achieved and studied. This polyatomic molecule has high sensitivity to physics beyond the Standard Model and represents a new class of species for future high precision probes of new T violating physics. The transverse temperature of the YbOH beam is reduced by nearly two orders of magnitude to 600 , mu K and the phase space density is increased by a factor of 6 via Sisyphus cooling. We develop a full numerical model of the laser cooling of YbOH and find excellent agreement with the data. We project that laser cooling and magneto optical trapping of long lived samples of YbOH molecules are within reach and these will allow a high sensitivity probe of the electric dipole moment (EDM) of the electron. The approach demonstrated here is easily generalized to other isotopologues of YbOH that have enhanced sensitivity to other symmetry violating electromagnetic moments.
[ 7 ]
physics.comp-ph
Deep learning inter atomic potential model for accurate irradiation damage simulations. We propose a hybrid scheme that interpolates smoothly the Ziegler Biersack Littmark (ZBL) screened nuclear repulsion potential with a newly developed deep learning potential energy model. The resulting DP ZBL model can not only provide overall good performance on the predictions of near equilibrium material properties but also capture the right physics when atoms are extremely close to each other, an event that frequently happens in computational simulations of irradiation damage events. We applied this scheme to the simulation of the irradiation damage processes in the face centered cubic aluminium system, and found better descriptions in terms of the defect formation energy, evolution of collision cascades, displacement threshold energy, and residual point defects, than the widely adopted ZBL modified embedded atom method potentials and its variants. Our work provides a reliable and feasible scheme to accurately simulate the irradiation damage processes and opens up new opportunities to solve the predicament of lacking accurate potentials for enormous newly discovered materials in the irradiation effect field.
[ 65, 19 ]
cs.CR cs.NI
Anonymity Network Tor and Performance Analysis of ARANEA an IOT Based Privacy Preserving Router. There was a time when the word security was only confined to the physical protection of things that were valuable which must be guarded against all the odds. Today, in a world where people can do things virtually have emerged the necessity to protect the virtual world. Every single facet of our life is being controlled by the internet one way or another. There is no privacy in the cyberspace as the data which we are browsing on the internet is being monitored on the other side by someone. Each work we are doing on the internet is getting tracked or the data are getting leaked without consent. To browse the internet securely we developed a router named Aranea which relates to the browser Tor. Tor gives traffic anonymity and security. The Tor browser can be used in both positive and negative purpose. Tor encrypts data, it hides the location and identity of the user, it hides the IP address of the device, it hides the network traffic and many more. By using Tor browser each user can browse the internet safely in the cyber world. Our goal is to create an additional security bridge through the router Aranea for every user so that each user can simply browse the internet anonymously.
[ 3 ]
cs.CL
ApplicaAI at SemEval 2020 Task 11 On RoBERTa CRF, Span CLS and Whether Self Training Helps Them. This paper presents the winning system for the propaganda Technique Classification (TC) task and the second placed system for the propaganda Span Identification (SI) task. The purpose of TC task was to identify an applied propaganda technique given propaganda text fragment. The goal of SI task was to find specific text fragments which contain at least one propaganda technique. Both of the developed solutions used semi supervised learning technique of self training. Interestingly, although CRF is barely used with transformer based language models, the SI task was approached with RoBERTa CRF architecture. An ensemble of RoBERTa based models was proposed for the TC task, with one of them making use of Span CLS layers we introduce in the present paper. In addition to describing the submitted systems, an impact of architectural decisions and training schemes is investigated along with remarks regarding training models of the same or better quality with lower computational budget. Finally, the results of error analysis are presented.
[ 46 ]
math.AG
Boundedness of log pluricanonical maps for surfaces of log general types in positive characteristic. In this article we prove the following boundedness result Fix a DCC set I subset 0, 1 . Let mathfrak D be the set of all log pairs (X, Delta) satisfying the following properties (i) X is a projective surface defined over an algebraically closed field, (ii) (X, Delta) is log canonical and the coefficients of Delta are in I , and (iii) K X Delta is big. Then there is a positive integer N N(I) depending only on the set I such that the linear system lceil m(K X Delta) rceil defines a birational map onto its image for all m geq N and (X, Delta) in mathfrak D
[ 67 ]
physics.atom-ph
Radiative QED corrections to one photon transition rates in hydrogen atom at finite temperatures. Within the framework of QED theory at finite temperature the thermal radiative corrections to spontaneous one photon transition rates in hydrogen atom are investigated. The radiative one loop self energy corrections are described in the thermal case. Closed analytical expressions are derived and their numerical calculations for the spontaneous decay rate of the 2p state are carried out. Dominance of thermal radiative corrections to spontaneous Ly alpha decay rate over ordinary induced transition rate up to temperatures T 6000 K is demonstrated.
[ 0 ]
cs.CV
Detector Algorithms of Bounding Box and Segmentation Mask of a Mask R CNN Model. Detection performances on bounding box and segmentation mask outputs of Mask R CNN models are evaluated. There are significant differences in detection performances of bounding boxes and segmentation masks, where the former is constantly superior to the latter. Harmonic values of precisions and recalls of linear cracks, joints, fillings, and shadows are significantly lower in segmentation masks than bounding boxes. Other classes showed similar harmonic values. Discussions are made on different performances of detection metrics of bounding boxes and segmentation masks focusing on detection algorithms of both detectors.
[ 32, 49, 1, 25 ]
math.OC cs.DS cs.LG math.PR
Online Allocation and Pricing Constant Regret via Bellman Inequalities. We develop a framework for designing simple and efficient policies for a family of online allocation and pricing problems, that includes online packing, budget constrained probing, dynamic pricing, and online contextual bandits with knapsacks. In each case, we evaluate the performance of our policies in terms of their regret (i.e., additive gap) relative to an offline controller that is endowed with more information than the online controller. Our framework is based on Bellman Inequalities, which decompose the loss of an algorithm into two distinct sources of error (1) arising from computational tractability issues, and (2) arising from estimation prediction of random trajectories. Balancing these errors guides the choice of benchmarks, and leads to policies that are both tractable and have strong performance guarantees. In particular, in all our examples, we demonstrate constant regret policies that only require re solving an LP in each period, followed by a simple greedy action selection rule thus, our policies are practical as well as provably near optimal.
[ 49, 28, 57 ]
math.OC cs.CG cs.PL
An Efficient Parametric Linear Programming Solver and Application to Polyhedral Projection. Polyhedral projection is a main operation of the polyhedron abstract domain.It can be computed via parametric linear programming (PLP), which is more efficient than the classic Fourier Motzkin elimination method.In prior work, PLP was done in arbitrary precision rational arithmetic.In this paper, we present an approach where most of the computation is performed in floating point arithmetic, then exact rational results are reconstructed.We also propose a workaround for a difficulty that plagued previous attempts at using PLP for computations on polyhedra in general the linear programming problems are degenerate, resulting in redundant computations and geometric descriptions.
[ 15 ]
cs.IT math.IT
Polynomial Invariant Theory and Shape Enumerator of Self Dual Codes in the NRT Metric. In this paper we consider self dual NRT codes, that is, self dual codes in the metric space endowed with the Niederreiter Rosenbloom Tsfasman (NRT metric). We use polynomial invariant theory to describe the shape enumerator of a binary self dual, doubly even self dual, and doubly doubly even self dual NRT code C subseteq M n,2 ( mathbb F 2 ) . Motivated by these results we describe the number of invariant polinomials that we must find to describe the shape enumerator of a self dual NRT code of M n,s ( mathbb F 2 ) . We define the ordered flip of a matrix A in M k,ns ( mathbb F q ) and present some constructions of self dual NRT codes over mathbb F q . We further give an application of ordered flip to the classification of bidimensional self dual NRT codes.
[ 68, 21 ]
math.CO cs.DM
Compression with wildcards All exact, or all minimal hitting sets. Our main objective is the COMPRESSED enumeration (based on wildcards) of all minimal hitting sets of general hypergraphs. To the author's best knowledge the only previous attempt towards compression, due to Toda T , is based on BDD's and much different from our techniques. Numerical experiments show that traditional one by one enumeration schemes cannot compete against compressed enumeration when the degree of compression is high. Our method works particularly well in these two cases Either compressing all exact hitting sets, or all minimum cardinality hitting sets. It also supports parallelization and cut off (i.e. restriction to all minimal hitting sets of cardinality at most m).
[ 0 ]
cs.CV
Radius Adaptive Convolutional Neural Network. Convolutional neural network (CNN) is widely used in computer vision applications. In the networks that deal with images, CNNs are the most time consuming layer of the networks. Usually, the solution to address the computation cost is to decrease the number of trainable parameters. This solution usually comes with the cost of dropping the accuracy. Another problem with this technique is that usually the cost of memory access is not taken into account which results in insignificant speedup gain. The number of operations and memory access in a standard convolution layer is independent of the input content, which makes it limited for certain accelerations. We propose a simple modification to a standard convolution to bridge this gap. We propose an adaptive convolution that adopts different kernel sizes (or radii) based on the content. The network can learn and select the proper radius based on the input content in a soft decision manner. Our proposed radius adaptive convolutional neural network (RACNN) has a similar number of weights to a standard one, yet, results show it can reach higher speeds. The code has been made available at https github.com meisamrf racnn.
[ 25 ]
math.PR
Shotgun Assembly of Erdos Renyi Random Graphs. Graph shotgun assembly refers to the problem of reconstructing a graph from a collection of local neighborhoods. In this paper, we consider shotgun assembly of ER random graphs G(n, p n) , where p n n alpha for 0 alpha 1 . We consider both reconstruction up to isomorphism as well as exact reconstruction (recovering the vertex labels as well as the structure). We show that given the collection of distance 1 neighborhoods, G is exactly reconstructable for 0 alpha frac 1 3 , but not reconstructable for frac 1 2 alpha 1 . Given the collection of distance 2 neighborhoods, G is exactly reconstructable for alpha in left(0, frac 1 2 right) cup left( frac 1 2 , frac 3 5 right) , but not reconstructable for frac 3 4 alpha 1 .
[ 36, 7 ]
physics.chem-ph physics.comp-ph
Bypassing the energy functional in density functional theory Direct calculation of electronic energies from conditional probability densities. Density functional calculations can fail for want of an accurate exchange correlation approximation. The energy can instead be extracted from a sequence of density functional calculations of conditional probabilities (CP DFT). Simple CP approximations yield usefully accurate results for two electron ions, the hydrogen dimer, and the uniform gas at all temperatures. CP DFT has no self interaction error for one electron, and correctly dissociates H2, both major challenges. For warm dense matter, classical CP DFT calculations can overcome the convergence problems of Kohn Sham DFT.
[ 23 ]
math.FA
Higher order weighted Sobolev spaces on the real line for strongly degenerate weights. Application to variational problems in elasticity of beams. For one dimensional interval and integrable weight function w we define via completion a weighted Sobolev space H m,p mu w of arbitrary integer order m . The weights in consideration may suffer strong degeneration so that, in general, functions u from H m,p mu w do not have weak derivatives. This contribution is focussed on studying the continuity properties of functions u at a chosen internal point x 0 to which we attribute a notion of criticality of order k and with respect to the weight w . For non critical points x 0 we formulate a local embedding result that guarantees continuity of functions u or their derivatives. Conversely, we employ duality theory to show that criticality of x 0 furnishes a smooth approximation of functions in H m,p mu w admitting jump type discontinuities at x 0 . The work concludes with demonstration of established results in the context of variational problem in elasticity theory of beams with degenerate width distribution.
[ 12, 21, 38 ]
math.NT math.CA math.CO
On the density of sumsets. Recently introduced by the authors in Proc. Edinb. Math. Soc. 60 (2020), 139 167 , quasi densities form a large family of real valued functions partially defined on the power set of the integers that serve as a unifying framework for the study of many known densities (including the asymptotic density, the Banach density, the logarithmic density, the analytic density, and the P 'olya density). We further contribute to this line of research by proving that (i) for each n in mathbf N and alpha in 0,1 , there is A subseteq mathbf N with kA in text dom ( mu) and mu(kA) alpha k n for every quasi density mu and every k 1, ldots, n , where kA A cdots A is the k fold sumset of A and text dom ( mu) denotes the domain of definition of mu (ii) for each alpha in 0,1 and every non empty finite B subseteq mathbf N , there is A subseteq mathbf N with A B in mathrm dom ( mu) and mu(A B) alpha for every quasi density mu (iii) for each alpha in 0,1 , there exists A subseteq mathbf N with 2A mathbf N such that A in text dom ( mu) and mu(A) alpha for every quasi density mu . Proofs rely on the properties of a little known density first considered by R.C. Buck and the "structure" of the set of all quasi densities in particular, they are rather different than previously known proofs of special cases of the same results.
[ 21 ]
math.CO
Biembeddings of cycle systems using integer Heffter arrays. In this paper we will show the existence of a face 2 colourable biembedding of the complete graph onto an orientable surface where each face is a cycle of a fixed length k , for infinitely many values of k . In particular, under certain conditions, we show that there exists at least (n 2) (p 2)! 2 (e 2 kn) non isomorphic face 2 colourable biembeddings of K 2nk 1 in which all faces are cycles of length k 4p 3 . These conditions are n equiv 1 mod 4 , k equiv 3 mod 4 and either n is prime or n gg k and n equiv 0 mod 3 implies p equiv 1 mod 3 . To achieve this result we begin by verifying the existence of (n 2) (p 2)! e 2 non equivalent Heffter arrays, H(n k) , which satisfy the conditions (1) for each row and each column the sequential partial sums determined by the natural ordering must be distinct modulo 2nk 1 (2) the composition of the natural orderings of the rows and columns is equivalent to a single cycle permutation on the entries in the array. The existence of Heffter arrays H(n k) that satisfy condition (1) was established earlier in cite BCDY and in this current paper we vary this construction and show that there are at least (n 2) (p 2)! e 2 such non equivalent H(n k) that satisfy condition (1) and then show that each of these Heffter arrays also satisfy condition (2) under certain conditions.
[ 1 ]
cs.LG
Logic Synthesis Meets Machine Learning Trading Exactness for Generalization. Logic synthesis is a fundamental step in hardware design whose goal is to find structural representations of Boolean functions while minimizing delay and area. If the function is completely specified, the implementation accurately represents the function. If the function is incompletely specified, the implementation has to be true only on the care set. While most of the algorithms in logic synthesis rely on SAT and Boolean methods to exactly implement the care set, we investigate learning in logic synthesis, attempting to trade exactness for generalization. This work is directly related to machine learning where the care set is the training set and the implementation is expected to generalize on a validation set. We present learning incompletely specified functions based on the results of a competition conducted at IWLS 2020. The goal of the competition was to implement 100 functions given by a set of care minterms for training, while testing the implementation using a set of validation minterms sampled from the same function. We make this benchmark suite available and offer a detailed comparative analysis of the different approaches to learning
[ 49, 50 ]
physics.med-ph math.OC
On Adaptation Cost and Tractability in Robust Adaptive Radiation Therapy Optimization. A framework for online robust adaptive radiation therapy (ART) is presented. This framework is designed to (i) handle interfractional geometric variations following a probability distribution different from the a priori hypothesis, (ii) address adaptation cost and (iii) address computational tractability. The novelty of this framework is the use of Bayesian inference and scenario reduction, which is evaluated in a series of treatment on a one dimensional phantom geometry. The initial robust plan is generated from a robust optimization problem based on either expected value or worst case optimization approach using the a priori hypothesis of the probability distribution governing the interfractional geometric variations. The actual interfractional variations are evaluated in terms of their likelihood with respect to the a priori hypothesis and violation of user specified tolerance limits. During an adaptation the a posteriori distribution is computed from the actual variations using Bayesian inference. The adapted plan is optimized to better suit the actual interfractional variations of the individual case, which is used until the next adaptation is. To address adaptation cost, the proposed framework provides an option for increased adaptation frequency. Computational tractability is addressed by scenario reduction algorithms to reduce the size of the optimization problem. According to the simulations, the proposed framework may improve target coverage compared to the corresponding non adaptive robust approach. Combining the worst case optimization approach with Bayesian inference may perform best in terms of improving CTV coverage and organ at risk (OAR) protection. Bayesian inference may have a greater impact on handling adaptation cost than increased adaptation frequency. The concept of scenario reduction may be useful to address computational tractability in ART and robust planning.
[ 23 ]
math.FA
Algebraic reflexivity of diameter preserving linear bijections between C(X) spaces. We prove that if X and Y are first countable compact Hausdorff spaces, then the set of all diameter preserving linear bijections from C(X) to C(Y) is algebraically reflexive.
[ 0 ]
cs.CV
ICDAR2019 Robust Reading Challenge on Multi lingual Scene Text Detection and Recognition RRC MLT 2019. With the growing cosmopolitan culture of modern cities, the need of robust Multi Lingual scene Text (MLT) detection and recognition systems has never been more immense. With the goal to systematically benchmark and push the state of the art forward, the proposed competition builds on top of the RRC MLT 2017 with an additional end to end task, an additional language in the real images dataset, a large scale multi lingual synthetic dataset to assist the training, and a baseline End to End recognition method. The real dataset consists of 20,000 images containing text from 10 languages. The challenge has 4 tasks covering various aspects of multi lingual scene text (a) text detection, (b) cropped word script classification, (c) joint text detection and script classification and (d) end to end detection and recognition. In total, the competition received 60 submissions from the research and industrial communities. This paper presents the dataset, the tasks and the findings of the presented RRC MLT 2019 challenge.
[ 78 ]
math.AC
On 1 absorbing primary ideal of a commutative ring (Correction to Theorem 17 is added). Let R be a commutative ring with identity. In this paper, we introduce the concept 1 absrbing primary ideal of R.
[ 19 ]
cs.CR
Towards usable automated detection of CPU architecture and endianness for arbitrary binary files and object code sequences. Static and dynamic binary analysis techniques are actively used to reverse engineer software's behavior and to detect its vulnerabilities, even when only the binary code is available for analysis. To avoid analysis errors due to misreading op codes for a wrong CPU architecture, these analysis tools must precisely identify the Instruction Set Architecture (ISA) of the object code under analysis. The variety of CPU architectures that modern security and reverse engineering tools must support is ever increasing due to massive proliferation of IoT devices and the diversity of firmware and malware targeting those devices. Recent studies concluded that falsely identifying the binary code's ISA caused alone about 10 of failures of IoT firmware analysis. The state of the art approaches to detect ISA for arbitrary object code look promising their results demonstrate effectiveness and high performance. However, they lack the support of publicly available datasets and toolsets, which makes the evaluation, comparison, and improvement of those techniques, datasets, and machine learning models quite challenging (if not impossible). This paper bridges multiple gaps in the field of automated and precise identification of architecture and endianness of binary files and object code. We develop from scratch the toolset and datasets that are lacking in this research space. As such, we contribute a comprehensive collection of open data, open source, and open API web services. We also attempt experiment reconstruction and cross validation of effectiveness, efficiency, and results of the state of the art methods. When training and testing classifiers using solely code sections from executable binary files, all our classifiers performed equally well achieving over 98 accuracy. The results are consistent and comparable with the current state of the art, hence supports the general validity of the algorithms
[ 8, 11 ]
physics.app-ph physics.optics
Plasmonic sensing with FIB milled 2D micro arrays of truncated gold nano pyramids. Plasmonic platforms are a promising solution for the next generation of low cost, integrated biomedical sensors. However, fabricating these nanostructures often requires lengthy and challenging fabrication processes. Here we exploit the peculiar features of Focused Ion Beam milling to obtain single step patterning of a plasmonic two dimensional (2D) array of truncated gold nano pyramids (TNP), with gaps as small as 17 nm. We describe the formation of plasmonic bandgaps in the arrangement of crossed tapered grooves that separate the nano pyramids, and we demonstrate refractive index sensing via dark field imaging in patterned areas of 30 mu m times 30 mu m.
[ 1, 3 ]
cs.CL cs.LG
AvgOut A Simple Output Probability Measure to Eliminate Dull Responses. Many sequence to sequence dialogue models tend to generate safe, uninformative responses. There have been various useful efforts on trying to eliminate them. However, these approaches either improve decoding algorithms during inference, rely on hand crafted features, or employ complex models. In our work, we build dialogue models that are dynamically aware of what utterances or tokens are dull without any feature engineering. Specifically, we start with a simple yet effective automatic metric, AvgOut, which calculates the average output probability distribution of all time steps on the decoder side during training. This metric directly estimates which tokens are more likely to be generated, thus making it a faithful evaluation of the model diversity (i.e., for diverse models, the token probabilities should be more evenly distributed rather than peaked at a few dull tokens). We then leverage this novel metric to propose three models that promote diversity without losing relevance. The first model, MinAvgOut, directly maximizes the diversity score through the output distributions of each batch the second model, Label Fine Tuning (LFT), prepends to the source sequence a label continuously scaled by the diversity score to control the diversity level the third model, RL, adopts Reinforcement Learning and treats the diversity score as a reward signal. Moreover, we experiment with a hybrid model by combining the loss terms of MinAvgOut and RL. All four models outperform their base LSTM RNN model on both diversity and relevance by a large margin, and are comparable to or better than competitive baselines (also verified via human evaluation). Moreover, our approaches are orthogonal to the base model, making them applicable as an add on to other emerging better dialogue models in the future.
[ 0 ]
cs.CV
An Improved Person Re identification Method by light weight convolutional neural network. Person Re identification is defined as a recognizing process where the person is observed by non overlapping cameras at different places. In the last decade, the rise in the applications and importance of Person Re identification for surveillance systems popularized this subject in different areas of computer vision. Person Re identification is faced with challenges such as low resolution, varying poses, illumination, background clutter, and occlusion, which could affect the result of recognizing process. The present paper aims to improve Person Re identification using transfer learning and application of verification loss function within the framework of Siamese network. The Siamese network receives image pairs as inputs and extract their features via a pre trained model. EfficientNet was employed to obtain discriminative features and reduce the demands for data. The advantages of verification loss were used in the network learning. Experiments showed that the proposed model performs better than state of the art methods on the CUHK01 dataset. For example, rank5 accuracies are 95.2 ( 5.7) for the CUHK01 datasets. It also achieved an acceptable percentage in Rank 1. Because of the small size of the pre trained model parameters, learning speeds up and there will be a need for less hardware and data.
[ 42, 76 ]
math.DG math.SP
On Monge Amp ere volumes of direct images. This paper is devoted to the study of the asymptotics of Monge Amp ere volumes of direct images associated with high tensor powers of an ample line bundle. We study the leading term of this asymptotics and provide a classification of bundles saturating the topological bound of Demailly. In the special case of high symmetric powers of ample vector bundles, this provides a characterization of vector bundles admitting projectively flat Hermitian structures.
[ 25, 23 ]
math.PR math.FA
Concentration of Markov chains with bounded moments. Let W t t 1 infty be a finite state stationary Markov chain, and suppose that f is a real valued function on the state space. If f is bounded, then Gillman's expander Chernoff bound (1993) provides concentration estimates for the random variable f(W 1) cdots f(W n) that depend on the spectral gap of the Markov chain and the assumed bound on f . Here we obtain analogous inequalities assuming only that the q 'th moment of f is bounded for some q geq 2 . Our proof relies on reasoning that differs substantially from the proofs of Gillman's theorem that are available in the literature, and it generalizes to yield dimension independent bounds for mappings f that take values in an L p( mu) for some p ge 2 , thus answering (even in the Hilbertian special case p 2 ) a question of Kargin (2007).
[ 17, 43 ]
cs.DB cs.DC
Delivery, consistency, and determinism rethinking guarantees in distributed stream processing. Consistency requirements for state of the art stream processing systems are defined in terms of delivery guarantees. Exactly once is the strongest one and the most desirable for end user. However, there are several issues regarding this concept. Commonly used techniques that enforce exactly once produce significant performance overhead. Besides, the notion of exactly once is not formally defined and does not capture all properties that provide stream processing systems supporting this guarantee. In this paper, we introduce a formal framework that allows us to define streaming guarantees more regularly. We demonstrate that the properties of delivery, consistency, and determinism are tightly connected within distributed stream processing. We also show that having lightweight determinism, it is possible to provide exactly once with almost no performance overhead. Experiments show that the proposed approach can significantly outperform alternative industrial solutions.
[ 0 ]
cs.CV
A Flexible Framework for Designing Trainable Priors with Adaptive Smoothing and Game Encoding. We introduce a general framework for designing and training neural network layers whose forward passes can be interpreted as solving non smooth convex optimization problems, and whose architectures are derived from an optimization algorithm. We focus on convex games, solved by local agents represented by the nodes of a graph and interacting through regularization functions. This approach is appealing for solving imaging problems, as it allows the use of classical image priors within deep models that are trainable end to end. The priors used in this presentation include variants of total variation, Laplacian regularization, bilateral filtering, sparse coding on learned dictionaries, and non local self similarities. Our models are fully interpretable as well as parameter and data efficient. Our experiments demonstrate their effectiveness on a large diversity of tasks ranging from image denoising and compressed sensing for fMRI to dense stereo matching.
[ 8 ]
physics.optics
Bragg soliton compression and fission on a CMOS compatible platform. Higher order soliton dynamics, specifically soliton compression and fission, underpin crucial applications in ultrafast optics, sensing, communications, and signal processing. Bragg solitons exploit the strong dispersive properties of periodic media near the photonic band edge, enabling soliton dynamics to occur on chip scale propagation distances and opening avenues to harness soliton compression and fission in integrated photonic platforms. However, implementation in CMOS compatible platforms has been hindered by the strong nonlinear loss that dominates the propagation of high intensity pulses in silicon and the low optical nonlinearity of traditional silicon nitride. Here, we present CMOS compatible, on chip Bragg solitons, with the largest soliton effect pulse compression to date with a factor of x5.7, along with the first time resolved measurements of soliton fission on a CMOS compatible platform. These observations were enabled by the combination of unique cladding modulated Bragg grating design, the high nonlinearity and negligible nonlinear loss of compositionally engineered ultra silicon rich nitride (USRN Si7N3).
[ 20 ]
math.NA cs.NA
An Integral Equation Method for the Cahn Hilliard Equation in the Wetting Problem. We present an integral equation approach to solving the Cahn Hilliard equation equipped with boundary conditions that model solid surfaces with prescribed Young's angles. The discretization of the system in time using convex splitting leads to a modified biharmonic equation at each time step. To solve it, we split the solution into a volume potential computed with free space kernels, plus the solution to a second kind integral equation (SKIE). The volume potential is evaluated with the help of a box based volume FMM method. For non box domains, source density is extended by solving a biharmonic Dirichlet problem. The near singular boundary integrals are computed using quadrature by expansion (QBX) with FMM acceleration. Our method has linear complexity in the number of surface volume degrees of freedom and can achieve high order convergence with adaptive refinement to manage error from function extension.
[ 21 ]
math.CO
Domination in direct products of complete graphs. Let X n denote the unitary Cayley graph of mathbb Z n mathbb Z . We continue the study of cases in which the inequality gamma t(X n) le g(n) is strict, where gamma t denotes the total domination number, and g is the arithmetic function known as Jacobsthal's function. The best that is currently known in this direction is a construction of Burcroff which gives a family of n with arbitrarily many prime factors that satisfy gamma t(X n) le g(n) 2 . We present a new interpretation of the problem which allows us to use recent results on the computation of Jacobsthal's function to construct n with arbitrarily many prime factors that satisfy gamma t(X n) le g(n) 16 . We also present new lower bounds on the domination numbers of direct products of complete graphs, which in turn allow us to derive new asymptotic lower bounds on gamma(X n) , where gamma denotes the domination number. Finally, resolving a question of Defant and Iyer, we completely classify all graphs G prod i 1 t K n i satisfying gamma(G) t 2 .
[ 11 ]
physics.app-ph
Versatile stabilized finite element formulations for nearly and fully incompressible solid mechanics. Computational formulations for large strain, polyconvex, nearly incompressible elasticity have been extensively studied, but research on enhancing solution schemes that offer better tradeoffs between accuracy, robustness, and computational efficiency remains to be highly relevant. In this paper, we present two methods to overcome locking phenomena, one based on a displacement pressure formulation using a stable finite element pairing with bubble functions, and another one using a simple pressure projection stabilized P1 P1 finite element pair. A key advantage is the versatility of the proposed methods with minor adjustments they are applicable to all kinds of finite elements and generalize easily to transient dynamics. The proposed methods are compared to and verified with standard benchmarks previously reported in the literature. Benchmark results demonstrate that both approaches provide a robust and computationally efficient way of simulating nearly and fully incompressible materials.
[ 10, 7 ]
physics.ao-ph physics.comp-ph
Modelling with Volna OP2 Towards tsunami threat reduction. Accurate and efficient tsunami modelling is essential for providing tsunami forecasts and hazard assessments. Volna OP2 is a finite volume solver of the nonlinear shallow water equations and its capabilities of producing both faster than real time ensembles and high resolution inundation studies are presented here. The code is massively parallelised and can utilise various high performance computing architectures. When an earthquake is detected there is always some uncertainty on the source parameters. Generating a faster than real time ensemble for maximum wave heights which captures this uncertainty would be of great benefit to tsunami warning centres. The 2003 Boumerdes earthquake (Algeria) acts as a test case for showing Volna OP2's ability at rapidly forecasting regional maximum wave heights. Drawing on various earthquake sources proposed in the literature and scaling the magnitudes to mimic uncertainty on the source, 20 separate earthquake realisations are simulated for 4 hours real time in 97s on two Nvidia V100 GPUs. Further a reduced ensemble of the Lisbon 1755 tsunami with an emphasis on the effects to the Irish coastline is presented. Where again various earthquake sources have been drawn from the literature and simulated on a regional scale. Finally, a pilot study which builds upon the reduced ensemble results investigates the inundation of a Lisbon tsunami on key sections of the Irish coastline. The results of this pilot study highlight that the inundation is constrained to low lying areas with maximum run up heights of approx 3.4m being found.
[ 32 ]
cs.DS
A simple certifying algorithm for 3 edge connectivity. A linear time certifying algorithm for 3 edge connectivity is presented. Given an undirected graph G, if G is 3 edge connected, the algorithm generates a construction sequence as a positive certificate for G. Otherwise, the algorithm decomposes G into its 3 edge connected components and at the same time generates a construction sequence for each connected component as well as the bridges and a cactus representation of the cut pairs in G. All of these are done by making only one pass over G using an innovative graph contraction technique. Moreover, the graph need not be 2 edge connected.
[ 84, 5 ]
physics.acc-ph physics.plasm-ph
Electron Sources from Plasmas. Relativistic electrons are easily generated by self injection when an intense laser drives a wakefield in a plasma, giving rise to wide electron energy distributions. Several mechanisms involving additional laser beams or different gas composition or distribution can be used to improve the electron beam quality. These mechanisms are introduced and discussed in the perspective of using laser driven electron sources as injectors for plasma accelerators.
[ 26, 47 ]
math.RT math.AT
Structure of Semi Continuous Q Tame Persistence Modules. Using a result by Chazal, Crawley Boevey and de Silva concerning radicals of persistence modules, we show that every lower semi continuous q tame persistence module can be decomposed as a direct sum of interval modules and that every upper semi continuous q tame persistence module can be decomposed as a product of interval modules.
[ 14 ]
cs.HC
Touch? Speech? or Touch and Speech? Investigating Multimodal Interaction for Visual Network Exploration and Analysis. Interaction plays a vital role during visual network exploration as users need to engage with both elements in the view (e.g., nodes, links) and interface controls (e.g., sliders, dropdown menus). Particularly as the size and complexity of a network grow, interactive displays supporting multimodal input (e.g., touch, speech, pen, gaze) exhibit the potential to facilitate fluid interaction during visual network exploration and analysis. While multimodal interaction with network visualization seems like a promising idea, many open questions remain. For instance, do users actually prefer multimodal input over unimodal input, and if so, why? Does it enable them to interact more naturally, or does having multiple modes of input confuse users? To answer such questions, we conducted a qualitative user study in the context of a network visualization tool, comparing speech and touch based unimodal interfaces to a multimodal interface combining the two. Our results confirm that participants strongly prefer multimodal input over unimodal input attributing their preference to 1) the freedom of expression, 2) the complementary nature of speech and touch, and 3) integrated interactions afforded by the combination of the two modalities. We also describe the interaction patterns participants employed to perform common network visualization operations and highlight themes for future multimodal network visualization systems to consider.
[ 25 ]
math.PR
Jackson network in a random environment strong approximation. We consider a Jackson network with regenerative input flows in which every server is subject to a random environment influence generating breakdowns and repairs. They occur in accordance with two independent sequences of i.i.d. random variables. We establish a theorem on the strong approximation of the vector of queue lengths by a reflected Brownian motion in positive orthant.
[ 0 ]
cs.CV
Building Robust Industrial Applicable Object Detection Models Using Transfer Learning and Single Pass Deep Learning Architectures. The uprising trend of deep learning in computer vision and artificial intelligence can simply not be ignored. On the most diverse tasks, from recognition and detection to segmentation, deep learning is able to obtain state of the art results, reaching top notch performance. In this paper we explore how deep convolutional neural networks dedicated to the task of object detection can improve our industrial oriented object detection pipelines, using state of the art open source deep learning frameworks, like Darknet. By using a deep learning architecture that integrates region proposals, classification and probability estimation in a single run, we aim at obtaining real time performance. We focus on reducing the needed amount of training data drastically by exploring transfer learning, while still maintaining a high average precision. Furthermore we apply these algorithms to two industrially relevant applications, one being the detection of promotion boards in eye tracking data and the other detecting and recognizing packages of warehouse products for augmented advertisements.
[ 0 ]
cs.CV
SML Semantic Meta learning for Few shot Semantic Segmentation. The significant amount of training data required for training Convolutional Neural Networks has become a bottleneck for applications like semantic segmentation. Few shot semantic segmentation algorithms address this problem, with an aim to achieve good performance in the low data regime, with few annotated training images. Recently, approaches based on class prototypes computed from available training data have achieved immense success for this task. In this work, we propose a novel meta learning framework, Semantic Meta Learning (SML) which incorporates class level semantic descriptions in the generated prototypes for this problem. In addition, we propose to use the well established technique, ridge regression, to not only bring in the class level semantic information, but also to effectively utilise the information available from multiple images present in the training data for prototype computation. This has a simple closed form solution, and thus can be implemented easily and efficiently. Extensive experiments on the benchmark PASCAL 5i dataset under different experimental settings show the effectiveness of the proposed framework.
[ 21 ]
math.CO
The Hopf algebra of generic rectangulations. A family of permutations called 2 clumped permutations forms a basis for a sub Hopf algebra of the Malvenuto Reutenauer Hopf algebra of permutations. The 2 clumped permutations are in bijection with certain decompositions of a square into rectangles, called generic rectangulations. Thus, we can describe the Hopf algebra of 2 clumped permutations using generic rectangulations (we call this isomorphic Hopf algebra the Hopf algebra of generic rectangulations). In this paper, we describe the cover relations in a lattice of generic rectangulations that is a lattice quotient of the right weak order on permutations. We then use this lattice to describe the product and coproduct operations in the Hopf algebra of generic rectangulations.
[ 13, 23 ]
math.DS math.FA
Inverse of mathcal U frequently hypercyclic operators. We show that there exists an invertible mathcal U frequently hypercyclic operator on ell p( mathbb N ) ( 1 le p infty ) whose inverse is not mathcal U frequently hypercyclic.
[ 42 ]
math.DG
On Einstein Lorentzian nilpotent Lie groups. In this paper, we study Lorentzian left invariant Einstein metrics on nilpotent Lie groups. We show that if the center of such Lie groups is degenerate then they are Ricci flat and their Lie algebras can be obtained by the double extension process from an abelian Euclidean Lie algebra. We show that all nilpotent Lie groups up to dimension 5 endowed with a Lorentzian Einstein left invariant metric have degenerate center and we use this fact to give a complete classification of these metrics. We show that if mathfrak g is the Lie algebra of a nilpotent Lie group endowed with a Lorentzian left invariant Einstein metric with non zero scalar curvature then the center Z( mathfrak g ) of mathfrak g is nondegenerate Euclidean, the derived ideal mathfrak g , mathfrak g is nondegenerate Lorentzian and Z( mathfrak g ) subset mathfrak g , mathfrak g . We give the first examples of Ricci flat Lorentzian nilpotent Lie algebra with nondegenerate center.
[ 1, 55 ]
cs.LG physics.geo-ph
Multi Constitutive Neural Network for Large Deformation Poromechanics Problem. In this paper, we study the problem of large strain consolidation in poromechanics with deep neural networks (DNN). Given different material properties and different loading conditions, the goal is to predict pore pressure and settlement. We propose a novel method "multi constitutive neural network" (MCNN) such that one model can solve several different constitutive laws. We introduce a one hot encoding vector as an additional input vector, which is used to label the constitutive law we wish to solve. Then we build a DNN which takes ( hat X , hat t ) as input along with a constitutive law label and outputs the corresponding solution. It is the first time, to our knowledge, that we can evaluate multi constitutive laws through only one training process while still obtaining good accuracies. We found that MCNN trained to solve multiple PDEs outperforms individual neural network solvers trained with PDE in some cases.