Datasets:

yyu

/

arxiv-attrprompt

like 1

math.AP

New Examples of Irreducible Local Diffusion of Hyperbolic PDE's. Local diffusion of strictly hyperbolic higher order PDE's with constant coefficients at all simple singularities of corresponding wavefronts can be explained and recognized by only two local geometrical features of these wavefronts. We radically disprove the obvious conjecture extending this fact to arbitrary singularities namely, we present examples of diffusion at all non simple singularity classes of generic wavefronts in odd dimensional spaces, which are not reducible to diffusion at simple singular points.

math.CO

On The Log Concavity of Polygonal Figurate Number Sequences. This paper presents the log concavity of the m gonal figurate number sequences. The author gives and proves the recurrence formula for m gonal figurate number sequences and its corresponding quotient sequences which are found to be bounded. Finally, the author also show that for m geq 3 , the sequence big S n(m) big n geq 1 of m gonal figurate numbers is a log concave.

math.DS math.GT

Ergodic decompositions of geometric measures on Anosov homogeneous spaces. Let G be a connected semisimple real algebraic group and Gamma a Zariski dense Anosov subgroup of G with respect to a minimal parabolic subgroup P . Let N be the maximal horospherical subgroup of G given by the unipotent radical of P . We describe the N ergodic decompositions of all Burger Roblin measures as well as the A ergodic decompositions of all Bowen Margulis Sullivan measures on Gamma backslash G . As a consequence, we obtain the following refinement of the main result of LO the space of all it non trivial N invariant ergodic and P circ quasi invariant Radon measures on Gamma backslash G , up to constant multiples, is homeomorphic to mathbb R text rank ,G 1 times 1, cdots, k where k is the number of P circ minimal subsets in Gamma backslash G .

cs.CC

Tightness of Sensitivity and Proximity Bounds for Integer Linear Programs. We consider ILPs, where each variable corresponds to an integral point within a polytope mathcal P , i. e., ILPs of the form min c top x mid sum p in mathcal P cap mathbb Z d x p p b, x in mathbb Z mathcal P cap mathbb Z d ge 0 . The distance between an optimal fractional solution and an optimal integral solution (called proximity) is an important measure. A classical result by Cook et al. (Math. Program., 1986) shows that it is at most Delta Theta(d) where Delta is the largest coefficient in the constraint matrix. Another important measure studies the change in an optimal solution if the right hand side b is replaced by another right hand side b' . The distance between an optimal solution x w.r.t. b and an optimal solution x' w.r.t. b' (called sensitivity) is similarly bounded, i. e., lVert b b' rVert 1 cdot Delta Theta(d) , also shown by Cook et al. Even after more than thirty years, these bounds are essentially the best known bounds for these measures. While some lower bounds are known for these measures, they either only work for very small values of Delta , require negative entries in the constraint matrix, or have fractional right hand sides. Hence, these lower bounds often do not correspond to instances from algorithmic problems. This work presents for each Delta 0 and each d 0 ILPs of the above type with non negative constraint matrices such that their proximity and sensitivity is at least Delta Theta(d) . Furthermore, these instances are closely related to instances of the Bin Packing problem as they form a subset of columns of the configuration ILP. We thereby show that the results of Cook et al. are indeed tight, even for instances arising naturally from problems in combinatorial optimization.

cs.CL

Best Practices for Data Efficient Modeling in NLG How to Train Production Ready Neural Models with Less Data. Natural language generation (NLG) is a critical component in conversational systems, owing to its role of formulating a correct and natural text response. Traditionally, NLG components have been deployed using template based solutions. Although neural network solutions recently developed in the research community have been shown to provide several benefits, deployment of such model based solutions has been challenging due to high latency, correctness issues, and high data needs. In this paper, we present approaches that have helped us deploy data efficient neural solutions for NLG in conversational systems to production. We describe a family of sampling and modeling techniques to attain production quality with light weight neural network models using only a fraction of the data that would be necessary otherwise, and show a thorough comparison between each. Our results show that domain complexity dictates the appropriate approach to achieve high data efficiency. Finally, we distill the lessons from our experimental findings into a list of best practices for production level NLG model development, and present them in a brief runbook. Importantly, the end products of all of the techniques are small sequence to sequence models (2Mb) that we can reliably deploy in production.

math.PR

Extinction probabilities in branching processes with countably many types a general framework. We consider Galton Watson branching processes with countable typeset mathcal X . We study the vectors bf q (A) (q x(A)) x in mathcal X recording the conditional probabilities of extinction in subsets of types A subseteq mathcal X , given that the type of the initial individual is x . We first investigate the location of the vectors bf q (A) in the set of fixed points of the progeny generating vector and prove that q x( x ) is larger than or equal to the x th entry of any fixed point, whenever it is different from 1. Next, we present equivalent conditions for q x(A) q x (B) for any initial type x and A,B subseteq mathcal X . Finally, we develop a general framework to characterise all emph distinct extinction probability vectors, and thereby to determine whether there are finitely many, countably many, or uncountably many distinct vectors. We illustrate our results with examples, and conclude with open questions.

math.DG math.AP

Sufficient criteria for obtaining Hardy inequalities on Finsler manifolds. We establish Hardy inequalities involving a weight function on complete, not necessarily reversible Finsler manifolds. We prove that the superharmonicity of the weight function provides a sufficient condition to obtain Hardy inequalities. Namely, if rho is a nonnegative function and boldsymbol Delta rho geq 0 in weak sense, where boldsymbol Delta is the Finsler Laplace operator defined by boldsymbol Delta rho mathrm div ( boldsymbol nabla rho) , then we obtain the generalization of some Riemannian Hardy inequalities given in D'Ambrosio and Dipierro (Ann. Inst. H. Poincar 'e, 2013). By extending the results obtained, we prove a weighted Caccioppoli type inequality, a Gagliardo Nirenberg inequality and a Heisenberg Pauli Weyl uncertainty principle on complete Finsler manifolds. Furthermore, we present some Hardy inequalities on Finsler Hadamard manifolds with finite reversibility constant, by defining the weight function with the help of the distance function. Finally, we extend a weighted Hardy inequality to a class of Finsler manifolds of bounded geometry.

cs.LO cs.FL math.LO

A duality theoretic view on limits of finite structures. A systematic theory of structural limits for finite models has been developed by Nesetril and Ossona de Mendez. It is based on the insight that the collection of finite structures can be embedded, via a map they call the Stone pairing, in a space of measures, where the desired limits can be computed. We show that a closely related but finer grained space of measures arises via Stone Priestley duality and the notion of types from model theory by enriching the expressive power of first order logic with certain probabilistic operators''. We provide a sound and complete calculus for this extended logic and expose the functorial nature of this construction. The consequences are two fold. On the one hand, we identify the logical gist of the theory of structural limits. On the other hand, our construction shows that the duality theoretic variant of the Stone pairing captures the adding of a layer of quantifiers, thus making a strong link to recent work on semiring quantifiers in logic on words. In the process, we identify the model theoretic notion of types as the unifying concept behind this link. These results contribute to bridging the strands of logic in computer science which focus on semantics and on more algorithmic and complexity related areas, respectively.

cs.CV cs.LG cs.RO

Learning a Geometric Representation for Data Efficient Depth Estimation via Gradient Field and Contrastive Loss. Estimating a depth map from a single RGB image has been investigated widely for localization, mapping, and 3 dimensional object detection. Recent studies on a single view depth estimation are mostly based on deep Convolutional neural Networks (ConvNets) which require a large amount of training data paired with densely annotated labels. Depth annotation tasks are both expensive and inefficient, so it is inevitable to leverage RGB images which can be collected very easily to boost the performance of ConvNets without depth labels. However, most self supervised learning algorithms are focused on capturing the semantic information of images to improve the performance in classification or object detection, not in depth estimation. In this paper, we show that existing self supervised methods do not perform well on depth estimation and propose a gradient based self supervised learning algorithm with momentum contrastive loss to help ConvNets extract the geometric information with unlabeled images. As a result, the network can estimate the depth map accurately with a relatively small amount of annotated data. To show that our method is independent of the model structure, we evaluate our method with two different monocular depth estimation algorithms. Our method outperforms the previous state of the art self supervised learning algorithms and shows the efficiency of labeled data in triple compared to random initialization on the NYU Depth v2 dataset.

math.NT math.AG

Free rational points on smooth hypersurfaces. Motivated by a recent question of Peyre, we apply the Hardy Littlewood circle method to count "sufficiently free" rational points of bounded height on arbitrary smooth projective hypersurfaces of low degree that are defined over the rational numbers.

cs.RO

Safe Grasping with a Force Controlled Soft Robotic Hand. Safe yet stable grasping requires a robotic hand to apply sufficient force on the object to immobilize it while keeping it from getting damaged. Soft robotic hands have been proposed for safe grasping due to their passive compliance, but even such a hand can crush objects if the applied force is too high. Thus for safe grasping, regulating the grasping force is of uttermost importance even with soft hands. In this work, we present a force controlled soft hand and use it to achieve safe grasping. To this end, resistive force and bend sensors are integrated in a soft hand, and a data driven calibration method is proposed to estimate contact interaction forces. Given the force readings, the pneumatic pressures are regulated using a proportional integral controller to achieve desired force. The controller is experimentally evaluated and benchmarked by grasping easily deformable objects such as plastic and paper cups without neither dropping nor deforming them. Together, the results demonstrate that our force controlled soft hand can grasp deformable objects in a safe yet stable manner.

math.AP

On a class of reaction diffusion equations with aggregation. In this paper, global in time existence and blow up results are shown for a reaction diffusion equation appearing in the theory of aggregation phenomena (including chemotaxis). Properties of the corresponding steady state problem are also presented. Moreover, the stability around constant equilibria and the non existence of non constant solutions are studied in certain cases.

cs.AI

Neighborhood Cognition Consistent Multi Agent Reinforcement Learning. Social psychology and real experiences show that cognitive consistency plays an important role to keep human society in order if people have a more consistent cognition about their environments, they are more likely to achieve better cooperation. Meanwhile, only cognitive consistency within a neighborhood matters because humans only interact directly with their neighbors. Inspired by these observations, we take the first step to introduce emph neighborhood cognitive consistency (NCC) into multi agent reinforcement learning (MARL). Our NCC design is quite general and can be easily combined with existing MARL methods. As examples, we propose neighborhood cognition consistent deep Q learning and Actor Critic to facilitate large scale multi agent cooperations. Extensive experiments on several challenging tasks (i.e., packet routing, wifi configuration, and Google football player control) justify the superior performance of our methods compared with state of the art MARL approaches.

math.AP

Wolff Type Potential Estimates for Stationary Stokes Systems with Dini BMO Coefficients. The pointwise gradient estimate for weak solution pairs to the stationary Stokes system with Dini BMO coefficients is established via the Havin Maz'ya Wolff type nonlinear potential of the nonhomogeneous term. In addition, we present a pointwise bound for the weak solutions under no extra regularity assumption on the coefficients.

math.RA

Morita Equivalence of Graph and Ultragraph Leavitt Path Algebras. The primary purpose of this thesis is to show every ultragraph Leavitt path algebra is Morita equivalent, as a ring, to a graph Leavitt path algebra. Takeshi Katsura, Paul Muhly, Aidan Sims, and Mark Tomforde showed every ultragraph C algebra is Morita equivalent, in the C sense, to a graph C algebra our result is an algebraic analog of this fact. Further, we will use our result to give an alternate proof for established conditions which guarantee the simplicity of an ultragraph Leavitt path algebra over a field.

math.OA math.GR math.RT

Polish groups of unitaries. We study the question of which Polish groups can be realized as subgroups of the unitary group of a separable infinite dimensional Hilbert space. We also show that for a separable unital C algebra A , the identity component mathcal U 0(A) of its unitary group has property (OB) of Rosendal (hence it also has property (FH)) if and only if the algebra has finite exponential length (e.g. if it has real rank zero), while in many cases the unitary group mathcal U (A) does not have property (T). On the other hand, the p unitary group mathcal U p(M, tau) where M is a properly infinite semifinite von Neumann algbera with separable predual, does not have property (FH) for any 1 le p infty . This in particular solves a problem left unanswered in the work of Pestov cite Pestov18 .

cs.CR cs.LG cs.NA math.NA

Identification of Metallic Objects using Spectral MPT Signatures Object Characterisation and Invariants. The early detection of terrorist threats, such as guns and knives, through improved metal detection, has the potential to reduce the number of attacks and improve public safety and security. To achieve this, there is considerable potential to use the fields applied and measured by a metal detector to discriminate between different shapes and different metals since, hidden within the field perturbation, is object characterisation information. The magnetic polarizability tensor (MPT) offers an economical characterisation of metallic objects that can be computed for different threat and non threat objects and has an established theoretical background, which shows that the induced voltage is a function of the hidden object's MPT coefficients. In this paper, we describe the additional characterisation information that measurements of the induced voltage over a range of frequencies offer compared to measurements at a single frequency. We call such object characterisations its MPT spectral signature. Then, we present a series of alternative rotational invariants for the purpose of classifying hidden objects using MPT spectral signatures. Finally, we include examples of computed MPT spectral signature characterisations of realistic threat and non threat objects that can be used to train machine learning algorithms for classification purposes.

cs.RO cs.MA

Multiple criteria decision making for lane change model. Simulation has long been an essential part of testing autonomous driving systems, but only recently has simulation been useful for building and training self driving vehicles. Vehicle behavioural models are necessary to simulate the interactions between robot cars. This paper proposed a new method to formalize the lane changing model in urban driving scenarios. We define human incentives from different perspectives, speed incentive, route change incentive, comfort incentive and courtesy incentive etc. We applied a decision theoretical tool, called Multi Criteria Decision Making (MCDM) to take these incentive policies into account. The strategy of combination is according to different driving style which varies for each driving. Thus a lane changing decision selection algorithm is proposed. Not only our method allows for varying the motivation of lane changing from the purely egoistic desire to a more courtesy concern, but also they can mimic drivers' state, inattentive or concentrate, which influences their driving Behaviour. We define some cost functions and calibrate the parameters with different scenarios of traffic data. Distinguishing driving styles are used to aggregate decision makers' assessments about various criteria weightings to obtain the action drivers desire most. Our result demonstrates the proposed method can produce varied lane changing behaviour. Unlike other lane changing models based on artificial intelligence methods, our model has more flexible controllability.

math.CV

Marcinkiewicz multipliers associated with the Kohn Laplacian on the Shilov boundary of the product domain in mathbb C 2n . Let M (k) , k 1,2, ldots, n , be the boundary of an unbounded polynomial domain Omega (k) of finite type in mathbb C 2 , and let Box b (k) be the Kohn Laplacian on M (k) . In this paper, we study multivariable spectral multipliers m( Box b (1) , ldots, Box b (n) ) acting on the Shilov boundary widetilde M M (1) times cdots times M (n) of the product domain Omega (1) times cdots times Omega (n) . We show that if a function F( lambda 1, ldots , lambda n) satisfies a Marcinkiewicz type differential condition, then the spectral multiplier operator m( Box b (1) , ldots, Box b (n) ) is a product Calder 'on Zygmund operator of Journ 'e type.

physics.soc-ph cs.SI

A critique of the Mean Field Approximation in preferential attachment networks. The Mean Field Approximation (MFA), or continuum method, is often used in courses on Networks to derive the degree distribution of preferential attachment networks. This method is simple and the outcome is close to the correct answer. However, this paper shows that the method is flawed in several aspects, leading to unresolvable contradictions. More importantly, the MFA is not explicitly derived from a mathematical model. An analysis of the implied model shows that it makes an approximation which is far from the truth and another one which can not be motivated in general. The success of the MFA for preferential attachment networks is therefore accidental and the method is not suitable for teaching undergraduates.

cs.HC

From Ideation to Implications Directions for the Internet of Things in the Home. In this paper we give a brief overview of our approaches and ongoing work for future directions of the Internet of Things (IoT) with a focus on the IoT in the home. We highlight some of our activities including tools and methods for an ideation driven approach as well as for an implications driven approach. We point to some findings of workshops and empirical field studies. We show examples for new classes of idiosyncratic IoT devices, how implications emerge by (mis)using sensor data and how users interacted with IoT systems in shared spaces.

cs.IT cs.NI math.IT

Fast Beam Training for IRS Assisted Multiuser Communications. In this letter, we consider an intelligent reflecting surface (IRS) assisted multiuser communication system, where an IRS is deployed to provide virtual line of sight (LoS) links between an access point (AP) and multiple users. We consider the practical codebook based IRS passive beamforming and study efficient design for IRS reflect beam training, which is challenging due to the large number of IRS reflecting elements. In contrast to the conventional single beam training, we propose a new multi beam training method by dividing the IRS reflecting elements into multiple sub arrays and designing their simultaneous multi beam steering over time. By simply comparing the received signal power over time, each user can detect its optimal IRS beam direction with a high probability, even without searching over all possible beam directions as the single beam training. Simulation results show that our proposed multi beam training significantly reduces the training time of conventional single beam training and yet achieves comparable IRS passive beamforming performance for data transmission.

q-bio.QM

High sensitivity COVID 19 group testing by digital PCR. Background Worldwide demand for SARS CoV 2 RT PCR testing is increasing as more countries are impacted by COVID 19 and as testing remains central to contain the spread of the disease, both in countries where the disease is emerging and in countries that are past the first wave but exposed to re emergence. Group testing has been proposed as a solution to expand testing capabilities but sensitivity concerns have limited its impact on the management of the pandemic. Digital PCR (RT dPCR) has been shown to be more sensitive than RT PCR and could help in this context. Methods We implemented RT dPCR based COVID 19 group testing on commercially available system and assay (Naica System from Stilla Technologies) and investigated the sensitivity of the method in real life conditions of a university hospital in Paris, France, in May 2020. We tested the protocol in a direct comparison with reference RT PCR testing on 448 samples split into groups of 3 sizes for RT dPCR analysis 56 groups of 8 samples, 28 groups of 16 samples and 14 groups of 32 samples. Results Individual RT PCR testing identified 25 positive samples. Using groups of 8, testing by RT dPCR identified 23 groups as positive, corresponding to 26 true positive samples including 2 samples not initially detected by individual RT PCR but confirmed positive by further RT PCR and RT dPCR investigation. For groups of 16, 15 groups tested positive, corresponding to 25 true positive samples identified. 100 concordance is found for groups of 32 but with limited data points.

cs.CL cs.AI cs.LG

A New Data Normalization Method to Improve Dialogue Generation by Minimizing Long Tail Effect. Recent neural models have shown significant progress in dialogue generation. Most generation models are based on language models. However, due to the Long Tail Phenomenon in linguistics, the trained models tend to generate words that appear frequently in training datasets, leading to a monotonous issue. To address this issue, we analyze a large corpus from Wikipedia and propose three frequency based data normalization methods. We conduct extensive experiments based on transformers and three datasets respectively collected from social media, subtitles, and the industrial application. Experimental results demonstrate significant improvements in diversity and informativeness (defined as the numbers of nouns and verbs) of generated responses. More specifically, the unigram and bigram diversity are increased by 2.6 12.6 and 2.2 18.9 on the three datasets, respectively. Moreover, the informativeness, i.e. the numbers of nouns and verbs, are increased by 4.0 7.0 and 1.4 12.1 , respectively. Additionally, the simplicity and effectiveness enable our methods to be adapted to different generation models without much extra computational cost.

cs.GT math.PR

Flexibility can hurt dynamic matching system performance. We study the performance of general dynamic matching models. This model is defined by a connected graph, where nodes represent the class of items and the edges the compatibilities between items. Items of different classes arrive one by one to the system according to a given probability distribution. Upon arrival, an item is matched with a compatible item according to the First Come First Served discipline and leave the system immediately, whereas it is enqueued with other items of the same class, if any. We show that such a model may exhibit a non intuitive behavior increasing the services ability by adding new edges in the matching graph may lead to a larger average population. This is similar to a Braess paradox. We first consider a quasicomplete graph with four nodes and we provide values of the probability distribution of the arrivals such that when we add an edge the mean number of items is larger. Then, we consider an arbitrary matching graph and we show sufficient conditions for the existence or non existence of this paradox. We conclude that the analog to the Braess paradox in matching models is given when specific independent sets are in saturation, i.e., the system is close to the stability condition.

cs.CY

Interoperable Framework to Enhance Citizen Services in the Kingdom of Bahrain. Citizen records are scattered between different state organizations. It wastes time, effort, and resources for both citizen and organization to collect, maintain, and update records to fulfill citizen services. Interoperability is a key element that enables seamless collaboration between different entities. It requires non conventional methods to overcome interoperability challenges such as lack of trust, centralization, and policy and technology differences. Blockchain is a disruptive technology with the potential to overcome these challenges. The technology designed to enable peer to peer transactions with elimination of intermediary in a trustless environment through the control of consensus mechanisms. This research aims to explore the status of interoperability in Bahrain, design an interoperable framework, and then test the validity of the framework by implementation of a prototype using blockchain technology. The research will be divided into four phases I Information collection, II Design and modeling the framework, III Implementation of a prototype, and Phase IV Measuring the performance of the prototype. This research is in progress and it is expected, once is it complete, to enhance the e government's plan in the Kingdom of Bahrain to provide better services to citizens and help in the transition from e government to seamless government, which will lead to sustainable citizen services. On the other hand, the findings of the study is expected to improve the social, economical, and environmental sustainability by the increase in process optimization, reduction of cost and complexity.

math.CA

Fractional Leibniz integral rules for Riemann Liouville and Caputo fractional derivatives and their applications. In recent years, the theory for Leibniz integral rule in the fractional sense has not been able to get substantial development. As an urgent problem to be solved, we study a Leibniz integral rule for Riemann Liouville and Caputo type differentiation operators with general fractional order of n 1 alpha leq n , n in mathbb N . A rule of fractional differentiation under integral sign with general order is necessary and applicable tool for verification by substitution for candidate solutions of inhomogeneous multi term fractional differential equations. We derive explicit analytical solutions of generalized Bagley Torvik equations in terms of recently defined bivariate Mittag Leffler type functions that based on fractional Green's function method and verified solutions by substitution in accordance by applying the fractional Leibniz integral rule. Furthermore, we study an oscillator equation as a special case of differential equations with multi orders via the Leibniz integral rule.

math.NT math.AG

Mass formula and Oort's conjecture for supersingular abelian threefolds. Using the theory of polarised flag type quotients, we determine mass formulae for all principally polarised supersingular abelian threefolds defined over an algebraically closed field k of characteristic p . We combine these results with computations of the automorphism groups to study Oort's conjecture we prove that every generic three dimensional principally polarised supersingular abelian variety over k of characteristic neq 2 has automorphism group pm 1 .

cs.CL

Learning to Recover from Multi Modality Errors for Non Autoregressive Neural Machine Translation. Non autoregressive neural machine translation (NAT) predicts the entire target sequence simultaneously and significantly accelerates inference process. However, NAT discards the dependency information in a sentence, and thus inevitably suffers from the multi modality problem the target tokens may be provided by different possible translations, often causing token repetitions or missing. To alleviate this problem, we propose a novel semi autoregressive model RecoverSAT in this work, which generates a translation as a sequence of segments. The segments are generated simultaneously while each segment is predicted token by token. By dynamically determining segment length and deleting repetitive segments, RecoverSAT is capable of recovering from repetitive and missing token errors. Experimental results on three widely used benchmark datasets show that our proposed model achieves more than 4 times speedup while maintaining comparable performance compared with the corresponding autoregressive model.

cs.CV cs.LG

Inducing Predictive Uncertainty Estimation for Face Recognition. Knowing when an output can be trusted is critical for reliably using face recognition systems. While there has been enormous effort in recent research on improving face verification performance, understanding when a model's predictions should or should not be trusted has received far less attention. Our goal is to assign a confidence score for a face image that reflects its quality in terms of recognizable information. To this end, we propose a method for generating image quality training data automatically from 'mated pairs' of face images, and use the generated data to train a lightweight Predictive Confidence Network, termed as PCNet, for estimating the confidence score of a face image. We systematically evaluate the usefulness of PCNet with its error versus reject performance, and demonstrate that it can be universally paired with and improve the robustness of any verification model. We describe three use cases on the public IJB C face verification benchmark (i) to improve 1 1 image based verification error rates by rejecting low quality face images (ii) to improve quality score based fusion performance on the 1 1 set based verification benchmark and (iii) its use as a quality measure for selecting high quality (unblurred, good lighting, more frontal) faces from a collection, e.g. for automatic enrolment or display.

cs.DM math.CO

(Theta, triangle) free and (even hole, K 4 ) free graphs. Part 2 bounds on treewidth. A em theta is a graph made of three internally vertex disjoint chordless paths P 1 a dots b , P 2 a dots b , P 3 a dots b of length at least 2 and such that no edges exist between the paths except the three edges incident to a and the three edges incident to b . A em pyramid is a graph made of three chordless paths P 1 a dots b 1 , P 2 a dots b 2 , P 3 a dots b 3 of length at least 1, two of which have length at least 2, vertex disjoint except at a , and such that b 1b 2b 3 is a triangle and no edges exist between the paths except those of the triangle and the three edges incident to a . An emph even hole is a chordless cycle of even length. For three non negative integers i leq j leq k , let S i,j,k be the tree with a vertex v , from which start three paths with i , j , and k edges respectively. We denote by K t the complete graph on t vertices. We prove that for all non negative integers i, j, k , the class of graphs that contain no theta, no K 3 , and no S i, j, k as induced subgraphs have bounded treewidth. We prove that for all non negative integers i, j, k, t , the class of graphs that contain no even hole, no pyramid, no K t , and no S i, j, k as induced subgraphs have bounded treewidth. To bound the treewidth, we prove that every graph of large treewidth must contain a large clique or a minimal separator of large cardinality.

math.NT math.CO

On the monotonicity of additive representation functions. We study the monotonicity behavior of three slightly differently defined additive representation functions (as initiated by Erd H o s, S ' a rk " o zy and S ' o s), answering one open question and another one partially, and give a slightly simpler proof for a result of Chen and Tang.

cs.CV

An Autonomous Approach to Measure Social Distances and Hygienic Practices during COVID 19 Pandemic in Public Open Spaces. Coronavirus has been spreading around the world since the end of 2019. The virus can cause acute respiratory syndrome, which can be lethal, and is easily transmitted between hosts. Most states have issued state at home executive orders, however, parks and other public open spaces have largely remained open and are seeing sharp increases in public use. Therefore, in order to ensure public safety, it is imperative for patrons of public open spaces to practice safe hygiene and take preventative measures. This work provides a scalable sensing approach to detect physical activities within public open spaces and monitor adherence to social distancing guidelines suggested by the US Centers for Disease Control and Prevention (CDC). A deep learning based computer vision sensing framework is designed to investigate the careful and proper utilization of parks and park facilities with hard surfaces (e.g. benches, fence poles, and trash cans) using video feeds from a pre installed surveillance camera network. The sensing framework consists of a CNN based object detector, a multi target tracker, a mapping module, and a group reasoning module. The experiments are carried out during the COVID 19 pandemic between March 2020 and May 2020 across several key locations at the Detroit Riverfront Parks in Detroit, Michigan. The sensing framework is validated by comparing automatic sensing results with manually labeled ground truth results. The proposed approach significantly improves the efficiency of providing spatial and temporal statistics of users in public open spaces by creating straightforward data visualizations for federal and state agencies. The results can also provide on time triggering information for an alarming or actuator system which can later be added to intervene inappropriate behavior during this pandemic.

math.NA cs.NA physics.comp-ph

Adaptive integration of nonlinear evolution equations on tensor manifolds. We develop new adaptive algorithms for temporal integration of nonlinear evolution equations on tensor manifolds. These algorithms, which we call step truncation methods, are based on performing one time step with a conventional time stepping scheme, followed by a truncation operation onto a tensor manifold. By selecting the rank of the tensor manifold adaptively to satisfy stability and accuracy requirements, we prove convergence of a wide range of step truncation methods, including explicit one step and multi step methods. These methods are very easy to implement as they rely only on arithmetic operations between tensors, which can be performed by efficient and scalable parallel algorithms. Adaptive step truncation methods can be used to compute numerical solutions of high dimensional PDEs, which have become central to many new areas of application such optimal mass transport, random dynamical systems, and mean field optimal control. Numerical applications are presented and discussed for a Fokker Planck equation with spatially dependent drift on a flat torus of dimension two and four.

math.CA

Anisotropic Gaussian approximation in L 2( mathbb R 2) . Let mathcal D be the dictionary of Gaussian mixtures the functions created by affine change of variables of a single Gaussian in n dimensions. mathcal D is used pervasively in scientific applications to a degree that practitioners often employ it as their default choice for representing their scientific object. Its use in applications hinges on the perception that this dictionary is large enough, and its members are local enough in space and frequency, to provide efficient approximation to "almost all objects of interest". However, and perhaps surprisingly, only a handful of concrete theoretical results are actually known on the ability to use Gaussian mixtures in lieu of mainstream representation systems. The present paper shows that, in 2D, Gaussian mixtures are effective in resolving anisotropic structures, too. In this setup, the "smoothness class" is comprised of 2D functions that are sparsely represented using curvelets. An algorithm for N term approximation from (a small subset of) mathcal D is presented, and the error bounds are then shown to be on par with the errors of N term curvelet approximation. The latter are optimal, essentially by definition. Our approach is based on providing effective approximation from mathcal D to the members of the curvelet system, mimicking the approach in arXiv 0802.2517 and arXiv 0911.2803 where the mother wavelets are approximated. When the error is measured in the 1 norm, this adaptation of the prior approach, combined with standard tools, yields the desired results. However, handling the 2 norm case is much more subtle and requires substantial new machinery in this case, the error analysis cannot be solely done on the space domain some of it has to be carried out on frequency. Since, on frequency, all members of mathcal D are centered at the origin, a delicate analysis for controlling the error there is needed.

cs.GT

Smooth Fictitious Play in N times 2 Potential Games. The paper shows that smooth fictitious play converges to a neighborhood of a pure strategy Nash equilibrium with probability 1 in almost all N times 2 ( N player, two action) potential games. The neighborhood of convergence may be made arbitrarily small by taking the smoothing parameter to zero. Simple proof techniques are furnished by considering regular potential games.

math.CT

Dirichlet Functors are Contravariant Polynomial Functors. Polynomial functors are sums of covariant representable functors from the category of sets to itself. They have a robust theory with many applications from operads and opetopes to combinatorial species. In this paper, we define a contravariant analogue of polynomial functors Dirichlet functors. We develop the basic theory of Dirichlet functors, and relate them to their covariant analogues.

math.AG math.NT

On mathsf G isoshtukas over function fields. In this paper we classify isogeny classes of global mathsf G shtukas over a smooth projective curve C mathbb F q (or equivalently sigma conjugacy classes in mathsf G ( mathsf F otimes mathbb F q overline mathbb F q ) where mathsf F is the field of rational functions of C ) by two invariants bar kappa, bar nu extending previous works of Kottwitz. This result can be applied to study points of moduli spaces of mathsf G shtukas and thus is helpful to calculate their cohomology.

math.OC

Towards Dynamic Pricing for Shared Mobility on Demand using Markov Decision Processes and Dynamic Programming. In a Shared Mobility on Demand Service (SMoDS), dynamic pricing plays an important role in the form of an incentive for the decision of the empowered passenger on the ride offer. Strategies for determining the dynamic tariff should be suitably designed so that the incurred demand and supply are balanced and therefore economic efficiency is achieved. In this manuscript, we formulate a discrete time Markov Decision Process (MDP) to determine the probability desired by the SMoDS platform corresponding to the acceptance rate of each empowered passenger at each state of the system. We use Estimated Waiting Time (EWT) as the metric for the balance between demand and supply, with the goal that EWT be regulated around a target value. We then develop a Dynamic Programming (DP) algorithm to derive the optimal policy of the MDP that regulates EWT around the target value. Computational experiments are conducted that demonstrate the regulation of EWT is effective, through various scenarios. The overall demonstration is carried out offline. The MDP formulation together with the DP algorithm can be utilized to an online determination of the dynamic tariff by integrating with our earlier works on Cumulative Prospect Theory based passenger behavioral modeling and the AltMin dynamic routing algorithm, and form the subject of future works.

cs.AI cs.CY cs.IR cs.IT math.IT

Measuring Diversity in Heterogeneous Information Networks. Diversity is a concept relevant to numerous domains of research varying from ecology, to information theory, and to economics, to cite a few. It is a notion that is steadily gaining attention in the information retrieval, network analysis, and artificial neural networks communities. While the use of diversity measures in network structured data counts a growing number of applications, no clear and comprehensive description is available for the different ways in which diversities can be measured. In this article, we develop a formal framework for the application of a large family of diversity measures to heterogeneous information networks (HINs), a flexible, widely used network data formalism. This extends the application of diversity measures, from systems of classifications and apportionments, to more complex relations that can be better modeled by networks. In doing so, we not only provide an effective organization of multiple practices from different domains, but also unearth new observables in systems modeled by heterogeneous information networks. We illustrate the pertinence of our approach by developing different applications related to various domains concerned by both diversity and networks. In particular, we illustrate the usefulness of these new proposed observables in the domains of recommender systems and social media studies, among other fields.

physics.med-ph cs.CE physics.comp-ph q-bio.QM

Lattice Boltzmann method for simulation of diffusion magnetic resonance imaging physics in multiphase tissue models. We report an implementation of the lattice Boltzmann method (LBM) to integrate the Bloch Torrey equation, which describes the evolution of the transverse magnetization vector and the fate of the signal of diffusion magnetic resonance imaging (dMRI). Motivated by the need to interpret dMRI experiments in biological tissues, and to offset the small time step limitation of classical LBM, a hybrid LBM scheme is introduced and implemented to solve the Bloch Torrey equation. A membrane boundary condition is presented which is able to accurately represent the effects of thin curvilinear membranes typically found in biological tissues. As implemented, the hybrid LBM scheme accommodates piece wise uniform transport, dMRI parameters, periodic and mirroring outer boundary conditions, and finite membrane permeabilities on non boundary conforming inner boundaries. By comparing with analytical solutions of limiting cases, we demonstrate that the hybrid LBM scheme is more accurate than the classical LBM scheme. The proposed explicit LBM scheme maintains second order spatial accuracy, stability, and first order temporal accuracy for a wide range of parameters. The parallel implementation of the hybrid LBM code in a multi CPU computer system, as well as on GPUs, is straightforward and efficient. Along with offering certain advantages over finite element or Monte Carlo schemes, the proposed hybrid LBM constitutes a flexible scheme that can by easily adapted to model more complex interfacial conditions and physics in heterogeneous multiphase tissue models and to accommodate sophisticated dMRI sequences.

cs.CV cs.LG cs.RO

Learning Orientation Distributions for Object Pose Estimation. For robots to operate robustly in the real world, they should be aware of their uncertainty. However, most methods for object pose estimation return a single point estimate of the object's pose. In this work, we propose two learned methods for estimating a distribution over an object's orientation. Our methods take into account both the inaccuracies in the pose estimation as well as the object symmetries. Our first method, which regresses from deep learned features to an isotropic Bingham distribution, gives the best performance for orientation distribution estimation for non symmetric objects. Our second method learns to compare deep features and generates a non parameteric histogram distribution. This method gives the best performance on objects with unknown symmetries, accurately modeling both symmetric and non symmetric objects, without any requirement of symmetry annotation. We show that both of these methods can be used to augment an existing pose estimator. Our evaluation compares our methods to a large number of baseline approaches for uncertainty estimation across a variety of different types of objects.

math.FA math.AP

Dispersion, spreading and sparsity of Gabor wave packets for metaplectic and Schr "odinger operators. Sparsity properties for phase space representations of several types of operators have been extensively studied in recent papers, including pseudodifferential, Fourier integral and metaplectic operators, with applications to time frequency analysis of Schr "odinger type evolution equations. It has been proved that such operators are approximately diagonalized by Gabor wave packets. While the latter are expected to undergo some spreading phenomenon, there is no record of this issue in the aforementioned results. In this paper we prove refined estimates for the Gabor matrix of metaplectic operators, also of generalized type, where sparsity, spreading and dispersive properties are all noticeable. We provide applications to the propagation of singularities for the Schr "odinger equation.

q-bio.PE

A pragmatic approach to account for individual risks to optimise health policy. Developing feasible strategies and setting realistic targets for disease prevention and control depends on representative models, whether conceptual, experimental, logistical or mathematical. Mathematical modelling was established in infectious diseases over a century ago, with the seminal works of Ross and others. Propelled by the discovery of etiological agents for infectious diseases, and Koch's postulates, models have focused on the complexities of pathogen transmission and evolution to understand and predict disease trends in greater depth. This has led to their adoption by policy makers however, as model informed policies are being implemented, the inaccuracies of some predictions are increasingly apparent, most notably their tendency to overestimate the impact of control interventions. Here, we discuss how these discrepancies could be explained by methodological limitations in capturing the effects of heterogeneity in real world systems. We suggest that improvements could derive from theory developed in demography to study variation in life expectancy and ageing. Using simulations, we illustrate the problem and its impact, and formulate a pragmatic way forward.

cs.CL cs.LG

SHAP values for Explaining CNN based Text Classification Models. Deep neural networks are increasingly used in natural language processing (NLP) models. However, the need to interpret and explain the results from complex algorithms are limiting their widespread adoption in regulated industries such as banking. There has been recent work on interpretability of machine learning algorithms with structured data. But there are only limited techniques for NLP applications where the problem is more challenging due to the size of the vocabulary, high dimensional nature, and the need to consider textual coherence and language structure. This paper develops a methodology to compute SHAP values for local explainability of CNN based text classification models. The approach is also extended to compute global scores to assess the importance of features. The results are illustrated on sentiment analysis of Amazon Electronic Review data.

math.CA math.FA

Hyponormal Toeplitz Operators on Weighted Bergman Spaces. We consider operators acting on a Hilbert space that can be written as the sum of a shift and a diagonal operator and determine when the operator is hyponormal. The condition is presented in terms of the norm of an explicit block Jacobi matrix. We apply this result to the Toeplitz operator with specific algebraic symbols acting on certain weighted Bergman spaces of the unit disk and determine when such operators are hyponormal.

physics.ao-ph

Transition paths of marine debris and the stability of the garbage patches. We used transition path theory (TPT) to infer "reactive" pathways of floating marine debris trajectories. The TPT analysis was applied on a pollution aware time homogeneous Markov chain model constructed from trajectories produced by satellite tracked undrogued buoys from the NOAA Global Drifter Program. The latter involved coping with the openness of the system in physical space, which further required an adaptation of the standard TPT setting. Directly connecting pollution sources along coastlines with garbage patches of varied strengths, the unveiled reactive pollution routes represent alternative targets for ocean cleanup efforts. Among our specific findings we highlight constraining a highly probable pollution source for the Great Pacific Garbage Patch characterizing the weakness of the Indian Ocean gyre as a trap for plastic waste and unveiling a tendency of the subtropical gyres to export garbage toward the coastlines rather than to other gyres in the event of anomalously intense winds.

cs.HC cs.CY

Why Should the Q method be Integrated Into the Design Science Research? A Systematic Mapping Study. The Q method has been utilized over time in various areas, including information systems. In this study, we used a systematic mapping to illustrate how the Q method was applied within Information Systems (IS) community and proposing towards the integration of Q method into the Design Sciences Research (DSR) process as a tool for future research DSR based IS studies. In this mapping study, we collected peer reviewed journals from Basket of Eight journals and the digital library of the Association for Information Systems (AIS). Then we grouped the publications according to the process of DSR, and different variables for preparing Q method from IS publications. We found that the potential of the Q methodology can be used to support each main research stage of DSR processes and can serve as the useful tool to evaluate a system in the IS topic of system analysis and design

math.AP

Kwak Transform and Inertial Manifolds revisited. The paper gives sharp spectral gap conditions for existence of inertial manifolds for abstract semilinear parabolic equations with non self adjoint leading part. Main attention is paid to the case where this leading part have Jordan cells which appear after applying the so called Kwak transform to various important equations such as 2D Navier Stokes equations, reaction diffusion advection systems, etc. The different forms of Kwak transforms and relations between them are also discussed.

math.CO math.PR

On bucket increasing trees, clustered increasing trees and increasing diamonds. In this work we analyze bucket increasing tree families. We introduce two simple stochastic growth processes, generating random bucket increasing trees of size n , complementing the earlier result of Mahmoud and Smythe for bucket recursive trees. On the combinatorial side, we define multilabelled generalizations of the tree families d ary increasing trees and generalized plane oriented recursive trees. Additionally, we introduce a clustering process for ordinary increasing trees and relate it to bucket increasing trees. We discuss in detail the bucket size two and present a bijection between such bucket increasing tree families and certain families of graphs called increasing diamonds, providing an explanation for phenomena observed by Bodini et al.

math.MG math.CO

Hexagon tilings of the plane that are not edge to edge. An irregular vertex in a tiling by polygons is a vertex of one tile and belongs to the interior of an edge of another tile. In this paper we show that for any integer k geq 3 , there exists a normal tiling of the Euclidean plane by convex hexagons of unit area with exactly k irregular vertices. Using the same approach we show that there are normal edge to edge tilings of the plane by hexagons of unit area and exactly k many n gons ( n 6 ) of unit area. A result of Akopyan yields an upper bound for k depending on the maximal diameter and minimum area of the tiles. Our result complements this with a lower bound for the extremal case, thus showing that Akopyan's bound is asymptotically tight.

math.RT math.CO

On the Shifted Littlewood Richardson Coefficients and Littlewood Richardson Coefficients. We give a new interpretation of the shifted Littlewood Richardson coefficients f lambda mu nu ( lambda, mu, nu are strict partitions). The coefficients g lambda mu which appear in the decomposition of Schur Q function Q lambda into the sum of Schur functions Q lambda 2 l( lambda) sum mu g lambda mu s mu can be considered as a special case of f lambda mu nu (here lambda is a strict partition of length l( lambda) ). We also give another description for g lambda mu as the cardinal of a subset of a set that counts Littlewood Richardson coefficients c mu t mu tilde lambda . This new point of view allows us to establish connections between g lambda mu and c mu t mu tilde lambda . More precisely, we prove that g lambda mu g lambda mu t , and g lambda mu leq c mu t mu tilde lambda . We conjecture that g lambda mu 2 leq c tilde lambda mu t mu and formulate some conjectures on our combinatorial models which would imply this inequality if it is valid.

physics.optics

Magnetoplasmonics in nanocavities Dark plasmons enhance magneto optics beyond the intrinsic limit of magnetoplasmonic nanoantennas. Enhancing magneto optical effects is crucial for size reduction of key photonic devices based on non reciprocal propagation of light and to enable active nanophotonics. We disclose a so far unexplored approach that exploits dark plasmons to produce an unprecedented amplification of magneto optical activity. We designed and fabricated non concentric magnetoplasmonic disk plasmonic ring resonator nanocavities supporting multipolar dark modes. The broken geometrical symmetry of the design enables coupling with free space light and hybridization of dark modes of the ring nanoresonator with the dipolar localized plasmon resonance of the magnetoplasmonic disk. Such hybridization generates a multipolar resonance that amplifies the magneto optical response of the nanocavity by 1 order of magnitude with respect to the maximum enhancement achievable by localized plasmons in bare magnetoplasmonic nanoantennas. This large amplification results from the peculiar and enhanced electrodynamic response of the nanocavity, yielding an intense magnetically activated radiant magneto optical dipole driven by the low radiant multipolar resonance. The concept proposed is general and, therefore, our results open a new path that can revitalize research and applications of magnetoplasmonics to active nanophotonics and flat optics.

math.OC

Fully Distributed Nash Equilibrium Seeking in N Cluster Games. Distributed optimization and Nash equilibrium (NE) seeking problems have drawn much attention in the control community recently. This paper studies a class of non cooperative games, known as N cluster game, which subsumes both cooperative and non cooperative nature among multiple agents in the two problems solving distributed optimization problem within the cluster, while playing a non cooperative game across the clusters. Moreover, we consider a partial decision information game setup, i.e., the agents do not have direct access to other agents' decisions, and hence need to communicate with each other through a directed graph whose associated adjacency matrix is assumed to be non doubly stochastic. To solve the N cluster game problem, we propose a fully distributed NE seeking algorithm by a synthesis of leader following consensus and gradient tracking, where the leader following consensus protocol is adopted to estimate the other agents' decisions and the gradient tracking method is employed to trace some weighted average of the gradient. Furthermore, the algorithm is equipped with uncoordinated constant step sizes, which allows the agents to choose their own preferred step sizes, instead of a uniform coordinated step size. We prove that all agents' decisions converge linearly to their corresponding NE so long as the largest step size and the heterogeneity of the step size are small. We verify the derived results through a numerical example in a Cournot competition game.

cs.LO

A Mechanised Semantics for HOL with Ad hoc Overloading. Isabelle HOL augments classical higher order logic with ad hoc overloading of constant definitions that is, one constant may have several definitions for non overlapping types. In this paper, we present a mechanised proof that HOL with ad hoc overloading is consistent. All our results have been formalised in the HOL4 theorem prover.

cs.CV cs.LG

Learning to Group A Bottom Up Framework for 3D Part Discovery in Unseen Categories. We address the problem of discovering 3D parts for objects in unseen categories. Being able to learn the geometry prior of parts and transfer this prior to unseen categories pose fundamental challenges on data driven shape segmentation approaches. Formulated as a contextual bandit problem, we propose a learning based agglomerative clustering framework which learns a grouping policy to progressively group small part proposals into bigger ones in a bottom up fashion. At the core of our approach is to restrict the local context for extracting part level features, which encourages the generalizability to unseen categories. On the large scale fine grained 3D part dataset, PartNet, we demonstrate that our method can transfer knowledge of parts learned from 3 training categories to 21 unseen testing categories without seeing any annotated samples. Quantitative comparisons against four shape segmentation baselines shows that our approach achieve the state of the art performance.

math.CO math.RT

Upper bounds on Kronecker coefficients with few rows. We present three different upper bounds for Kronecker coefficients g( lambda, mu, nu) in terms of Kostka numbers, contingency tables and Littlewood Richardson coefficients. We then give various examples, asymptotic applications, and compare them with existing lower bounds.

cs.CV cs.LG

Exploiting Context for Robustness to Label Noise in Active Learning. Several works in computer vision have demonstrated the effectiveness of active learning for adapting the recognition model when new unlabeled data becomes available. Most of these works consider that labels obtained from the annotator are correct. However, in a practical scenario, as the quality of the labels depends on the annotator, some of the labels might be wrong, which results in degraded recognition performance. In this paper, we address the problems of i) how a system can identify which of the queried labels are wrong and ii) how a multi class active learning system can be adapted to minimize the negative impact of label noise. Towards solving the problems, we propose a noisy label filtering based learning approach where the inter relationship (context) that is quite common in natural data is utilized to detect the wrong labels. We construct a graphical representation of the unlabeled data to encode these relationships and obtain new beliefs on the graph when noisy labels are available. Comparing the new beliefs with the prior relational information, we generate a dissimilarity score to detect the incorrect labels and update the recognition model with correct labels which result in better recognition performance. This is demonstrated in three different applications scene classification, activity classification, and document classification.

math.QA

Diagram automorphisms and canonical bases for quantum affine algebras. Let mathbf U q be the negative part of the quantum enveloping algebra associated to a simply laced Kac Moody Lie algebra mathfrak g , and underline mathbf U q the algebra corresponding to the fixed point subalgebra of mathfrak g obtained from a diagram automorphism sigma on mathfrak g . Let mathbf B sigma be the set of sigma fixed elements in the canonical basis of mathbf U q , and underline mathbf B the canonical basis of underline mathbf U q . Lusztig proved that there exists a canonical bijection mathbf B sigma simeq underline mathbf B based on his geometric construction of canonical bases. In this paper, we prove (the signed bases version of) this fact, in the case where mathfrak g is finite or affine type, in an elementary way, in the sense that we don't appeal to the geometric theory of canonical bases nor Kashiwara's theory of crystal bases. We also discuss the correspondence between PBW bases, by using a new type of PBW bases of mathbf U q obtained by Muthiah Tingley, which is a generalization of PBW bases constructed by Beck Nakajima.

cs.PL cs.CR cs.LO

Dynamic IFC Theorems for Free!. We show that noninterference and transparency, the key soundness theorems for dynamic IFC libraries, can be obtained "for free", as direct consequences of the more general parametricity theorem of type abstraction. This allows us to give very short soundness proofs for dynamic IFC libraries such as faceted values and LIO. Our proofs stay short even when fully mechanized for Agda implementations of the libraries in terms of type abstraction.

physics.med-ph q-bio.TO

Where did the tumor start? An inverse solver with sparse localization for tumor growth models. We present a numerical scheme for solving an inverse problem for parameter estimation in tumor growth models for glioblastomas, a form of aggressive primary brain tumor. The growth model is a reaction diffusion partial differential equation (PDE) for the tumor concentration. We use a PDE constrained optimization formulation for the inverse problem. The unknown parameters are the reaction coefficient (proliferation), the diffusion coefficient (infiltration), and the initial condition field for the tumor PDE. Segmentation of Magnetic Resonance Imaging (MRI) scans from a single time snapshot drive the inverse problem where segmented tumor regions serve as partial observations of the tumor concentration. The precise time relative to tumor initiation is unknown, which poses an additional difficulty for inversion. We perform a frozen coefficient spectral analysis and show that the inverse problem is severely ill posed. We introduce a biophysically motivated regularization on the tumor initial condition. In particular, we assume that the tumor starts at a few locations (enforced with a sparsity constraint) and that the initial condition magnitude in the maximum norm equals one. We solve the resulting optimization problem using an inexact quasi Newton method combined with a compressive sampling algorithm for the sparsity constraint. Our implementation uses PETSc and AccFFT libraries. We conduct numerical experiments on synthetic and clinical images to highlight the improved performance of our solver over an existing solver that uses a two norm regularization for the calibration parameters. The existing solver is unable to localize the initial condition. Our new solver can localize the initial condition and recover infiltration and proliferation. In clinical datasets (for which the ground truth is unknown), our solver results in qualitatively different solutions compared to the existing solver.

cs.CV cs.AI

Recalibration of Neural Networks for Point Cloud Analysis. Spatial and channel re calibration have become powerful concepts in computer vision. Their ability to capture long range dependencies is especially useful for those networks that extract local features, such as CNNs. While re calibration has been widely studied for image analysis, it has not yet been used on shape representations. In this work, we introduce re calibration modules on deep neural networks for 3D point clouds. We propose a set of re calibration blocks that extend Squeeze and Excitation blocks and that can be added to any network for 3D point cloud analysis that builds a global descriptor by hierarchically combining features from multiple local neighborhoods. We run two sets of experiments to validate our approach. First, we demonstrate the benefit and versatility of our proposed modules by incorporating them into three state of the art networks for 3D point cloud analysis PointNet , DGCNN, and RSCNN. We evaluate each network on two tasks object classification on ModelNet40, and object part segmentation on ShapeNet. Our results show an improvement of up to 1 in accuracy for ModelNet40 compared to the baseline method. In the second set of experiments, we investigate the benefits of re calibration blocks on Alzheimer's Disease (AD) diagnosis. Our results demonstrate that our proposed methods yield a 2 increase in accuracy for diagnosing AD and a 2.3 increase in concordance index for predicting AD onset with time to event analysis. Concluding, re calibration improves the accuracy of point cloud architectures, while only minimally increasing the number of parameters.

math.OA

Rokhlin type properties, approximate innerness and Z stability. We establish four results concerning connections between actions on separable C algebras with Rokhlin type properties and absorption of the Jiang Su algebra Z. For actions of residually finite groups or of the reals which have finite Rokhlin dimension with commuting towers, we show that if the action of any nontrivial group element is approximately inner then the C algebra acted upon is Z stable. Without the assumption on approximate innerness, we show that the crossed product has good divisibility properties under mild assumptions. We also establish an analogous result for the generalized tracial Rokhlin property and tracial versions of approximate innerness and Z absorption for actions of finite groups and of the integers. For actions of a single automorphism which have the Rokhlin property, we show that a condition which is strictly weaker than requiring that some power of the automorphism is approximately inner is sufficient to obtain that the crossed product absorbs Z even when the original algebra is not.

cs.DS cs.DM

Faster Heuristics for Graph Burning. Graph burning is a process of information spreading through the network by an agent in discrete steps. The problem is to find an optimal sequence of nodes which have to be given information so that the network is covered in least number of steps. Graph burning problem is NP Hard for which two approximation algorithms and a few heuristics have been proposed in the literature. In this work, we propose three heuristics, namely, Backbone Based Greedy Heuristic (BBGH), Improved Cutting Corners Heuristic (ICCH) and Component Based Recursive Heuristic (CBRH). These are mainly based on Eigenvector centrality measure. BBGH finds a backbone of the network and picks vertex to be burned greedily from the vertices of the backbone. ICCH is a shortest path based heuristic and picks vertex to burn greedily from best central nodes. The burning number problem on disconnected graphs is harder than on the connected graphs. For example, burning number problem is easy on a path where as it is NP Hard on disjoint paths. In practice, large networks are generally disconnected and moreover even if the input graph is connected, during the burning process the graph among the unburned vertices may be disconnected. For disconnected graphs, ordering of the components is crucial. Our CBRH works well on disconnected graphs as it prioritizes the components. All the heuristics have been implemented and tested on several bench mark networks including large networks of size more than 50 K nodes. The experimentation also includes comparison to the approximation algorithms. The advantages of our algorithms are that they are much simpler to implement and also several orders faster than the heuristics proposed in the literature.

math.HO

Some remarks on the first Hardy Littlewood conjecture. Starting from the first Hardy Littlewood conjecture some topics will be covered an empirical approach to the distribution of the twin primes in classes mod(10) and a simplified proof of the Bruns theorem . Finally, it will be explored an approach based on numerical analysis Monte Carlo Method and Low discrepancy Sequences will be used to prove the convergence of the conjecture to the expected values.

cs.CY

IIT Kanpur Consulting Group Using Machine Learning and Management Consulting for Social Good. The IIT Kanpur Consulting Group is one of the pioneering research groups in India which focuses on the applications of Machine Learning and Strategy Consulting for social good. The group has been working since 2018 to help social organizations, nonprofits, and government entities in India leverage better insights from their data, with a special emphasis on the healthcare, environmental, and agriculture sectors. The group has worked on critical social problems which India is facing including Polio recurrence, COVID 19, air pollution and agricultural crop damage. This position paper summarises the focus areas and relevant projects which the group has worked on since its establishment, and also highlights the group's plans for using machine learning to address social problems during the COVID 19 crisis.

physics.ins-det physics.optics

Estimation of angular sensitivity for X ray interferometers with multiple phase gratings. Recently, X ray interferometers with more than one phase grating have been developed for differential phase contrast (DPC) imaging. In this study, a novel framework is developed to predict such interferometers' angular sensitivity responses (the minimum detectable refraction angle). Experiments are performed on the dual and triple phase grating interferometers, separately. Measurements show strong consistency with the predicted sensitivity values. Using this new approach, the DPC imaging performance of X ray interferometers with multiple phase gratings can be further optimized for future biomedical applications.

math.OC

Optimal control of the principal coefficient in a scalar wave equation. We consider optimal control of the scalar wave equation where the control enters as a coefficient in the principal part. Adding a total variation penalty allows showing existence of optimal controls, which requires continuity results for the coefficient to solution mapping for discontinuous coefficients. We additionally consider a so called "multi bang" penalty that promotes controls taking on values pointwise almost everywhere from a specified discrete set. Under additional assumptions on the data, we derive an improved regularity result for the state, leading to optimality conditions that can be interpreted in an appropriate pointwise fashion. The numerical solution makes use of a stabilized finite element method and a nonlinear primal dual proximal splitting algorithm.

cs.CR

Method of Counteraction in Social Engineering on Information Activity Objectives. The article presents a study using attacks such as a fake access point and a phishing page. The previous publications on social engineering have been reviewed, statistics of break ups are analyzed and directions and mechanism of realization of attacks having elements of social engineering are analyzed. The data from the research in three different places were collected and analyzed and the content statistics were provided. For comparison, three categories of higher education institutions were chosen technical, humanitarian and mixed profiles. Since the research was conducted in educational institutions during the week, most students in the experiment and graduate students took part in the experiment. For each educational institution, a registration form template was created that mimicked the design of the main pages. Examples of hardware and software implementation of a typical stand for attack, data collection and analysis are given. In order to construct a test stand, widely available components were chosen to show how easy it is to carry out attacks of this kind without significant initial costs and special skills. The article provides statistics on the number of connections, permission to use the address of the e mail and password, as well as permission to automatically transfer service data to the browser (cookies). The statistics are processed using specially written algorithms. The proposed approaches to solving the problem of socio technical attacks can be used and implemented for operation on any objects of information activity. As a result of the experiments, it is clear that the awareness of users of even technical specialties is not enough, so one needs to pay particular attention to the development of methods for raising awareness of users and reducing the number of potential attacks on objects of information activity.

cs.CC cs.GT

Time inconsistent Planning Simple Motivation Is Hard to Find. With the introduction of the graph theoretic time inconsistent planning model due to Kleinberg and Oren, it has been possible to investigate the computational complexity of how a task designer best can support a present biased agent in completing the task. In this paper, we study the complexity of finding a choice reduction for the agent that is, how to remove edges and vertices from the task graph such that a present biased agent will remain motivated to reach his target even for a limited reward. While this problem is NP complete in general, this is not necessarily true for instances which occur in practice, or for solutions which are of interest to task designers. For instance, a task designer may desire to find the best task graph which is not too complicated. We therefore investigate the problem of finding simple motivating subgraphs. These are structures where the agent will modify his plan at most k times along the way. We quantify this simplicity in the time inconsistency model as a structural parameter The number of branching vertices (vertices with out degree at least 2 ) in a minimal motivating subgraph. Our results are as follows We give a linear algorithm for finding an optimal motivating path, i.e. when k 0 . On the negative side, we show that finding a simple motivating subgraph is NP complete even if we allow only a single branching vertex revealing that simple motivating subgraphs are indeed hard to find. However, we give a pseudo polynomial algorithm for the case when k is fixed and edge weights are rationals, which might be a reasonable assumption in practice.

physics.optics

Arbitrary decomposition of a Mueller matrix. Mueller polarimetry involves a variety of instruments and technologies whose importance and scope of applications are rapidly increasing. The exploitation of these powerful resources depends strongly on the mathematical models that underlie the analysis and interpretation of the measured Mueller matrices and, very particularly, on the theorems for their serial and parallel decompositions. In this letter, the most general formulation for the parallel decomposition of a Mueller matrix is presented, which overcomes certain critical limitations of the previous approaches. In addition, the results obtained lead to a generalization of the polarimetric subtraction procedure and allow for a formulation of the arbitrary decomposition that integrates, in a natural way, the passivity criterion.

physics.comp-ph

Treecode accelerated Green Iteration for Kohn Sham Density Functional Theory. We present a real space computational method called treecode accelerated Green Iteration (TAGI) for all electron Kohn Sham Density Functional Theory. TAGI is based on a reformulation of the Kohn Sham equations in which the eigenvalue problem in differential form is converted into a fixed point problem in integral form by convolution with the modified Helmholtz Green's function. In each self consistent field (SCF) iteration, the fixed points are computed by Green Iteration, where the discrete convolution sums are efficiently evaluated by a GPU accelerated barycentric Lagrange treecode. Other techniques used in TAGI include adaptive mesh refinement, Fej 'er quadrature, singularity subtraction, gradient free eigenvalue update, and Anderson mixing to accelerate convergence of the SCF and Green Iterations. Ground state energy computations of several atoms (Li, Be, O) and small molecules (H 2 , CO, C 6 H 6 ) demonstrate TAGI's ability to efficiently achieve chemical accuracy.

cs.CY

Community Based Service Ecosystem Evolution Analysis. The prosperity of services and the frequent interaction between services contribute to the formation of the service ecosystem. Service ecosystem is a complex dynamic system with continuous evolution. Service providers voluntarily or compulsorily participate in this evolutionary process and face great opportunities and challenges. Existing studies on service ecosystem evolution are more about facilitating programmers to use services and have achieved remarkable results. However, the exploration of service ecosystem evolution from the business level is still insufficient. To make up this deficiency, in this paper, we present a method for analyzing service ecosystem evolution patterns from the perspective of the service community. Firstly, we train a service community evolution prediction model based on the community evolution sequences. Secondly, we explain the prediction model, showing how different factors affect the evolution of the service community. Finally, using the interpretable predictions and prior knowledge, we present how to assist service providers in making business decisions. Experiments on real world data show that this work can indeed provide business level insights into service ecosystem evolution. Additionally, all the data and well documented code used in this paper have been fully open source.

cs.SI physics.soc-ph

Football tracking networks Beyond event based connectivity. We propose using Network Science as a complementary tool to analyze player and team behavior during a football match. Specifically, we introduce four kinds of networks based on different ways of interaction between players. Our approach's main novelty is to use tracking datasets to create football tracking networks, instead of constructing and analyzing the traditional networks based on events. In this way, we are able to capture player interactions that go beyond passes and introduce the concepts of (a) Ball Flow Networks, (b) Marking Networks, (c) Signed Proximity Networks and (d) Functional Coordination Networks. After defining the methodology for creating each kind of network, we show some examples using tracking datasets from four different matches of LaLiga Santander. Finally, we discuss some of the applications, limitations, and further improvements of football tracking networks.

cs.IT cs.CC cs.CR math.AG math.IT

Interactive Oracle Proofs of Proximity to Algebraic Geometry Codes. In this work, we initiate the study of proximity testing to Algebraic Geometry (AG) codes. An AG code C C( mathcal C , mathcal P , D) is a vector space associated to evaluations on mathcal P of functions in the Riemann Roch space L mathcal C (D) . The problem of testing proximity to an error correcting code C consists in distinguishing between the case where an input word, given as an oracle, belongs to C and the one where it is far from every codeword of C . AG codes are good candidates to construct short proof systems, but there exists no efficient proximity tests for them. We aim to fill this gap. We construct an Interactive Oracle Proof of Proximity (IOPP) for some families of AG codes by generalizing an IOPP for Reed Solomon codes introduced by Ben Sasson, Bentov, Horesh and Riabzev, known as the FRI protocol. We identify suitable requirements for designing efficient IOPP systems for AG codes. Our approach relies on Kani's result that splits the Riemann Roch space of any invariant divisor under a group action on a curve into several explicit Riemann Roch spaces on the quotient curve. Under some hypotheses, a proximity test to C can thus be reduced to one to a simpler code C' . Iterating this process thoroughly, we end up with a membership test to a code with significantly smaller length. In addition to proposing the first proximity test targeting AG codes, our IOPP admits quasilinear prover arithmetic complexity and sublinear verifier arithmetic complexity with constant soundness for meaningful classes of AG codes. As a concrete instantiation, we study AG codes on Kummer curves, which are potentially much longer than Reed Solomon codes. For this type of curves, we manage to extend our generic construction to reach a strictly linear proving time and a strictly logarithmic verification time.

math.AP

(S) convergence and approximation of oscillatory solutions in fluid dynamics. We propose a new concept of (S) convergence applicable to numerical methods as well as other consistent approximations of the Euler system in gas dynamics. (S) convergence, based on averaging in the spirit of Strong Law of Large Numbers, reflects the asymptotic properties of a given approximate sequence better than the standard description via Young measures. Similarity with the tools of ergodic theory is discussed.

physics.soc-ph

A model for the generation of road networks. As part of the effort undertaken to understand urban environments and their generation, we need to explore models that produce statistically valid configurations of roads. These sort of models will help us to derive plausible mechanisms for the spatial location of population. This task is of fundamental importance, as we need to create an experimental environment that allows us to disentangle the specificities of a spatial configuration from the ideal system. Creating statistically valid models of road networks along with models of city generation allows to average the effects of geometry bringing us one step closer to understanding urban environments. To completely understand road networks we need to be able to grasp what principles of economy do their growth entail. It is therefore of interest to explore the possible shape that a performance function would have for transportation systems. In this work, we tackle this issue by proposing a network generation model based on a single parameter alpha which is capable of creating any type of network from trees to quasi surfaces and which is shown to generate networks close to the real road networks under study. This is obtained through the definition of local and weighted versions of centrality measures. These centrality measures deal with distance decay effects and nodes having different masses. We set ourselves to determine the properties and different regimes of this alpha model and we lay out a definition for the performance of a network taking into account factors such as robustness, construction cost, congestion and distance. We obtain the optimal alpha from the analysis of the space of possible performance functions, giving an intuition on the self organisational properties of the original network.

math.CA

The Hardy Littlewood theorem for double Fourier Haar series from Lebesgue spaces L bar p 0,1 with mixed metric and from net spaces N bar p , bar q (M) . In terms of the Fourier Haar coefficients, a criterion is obtained for the function f (x 1,x 2) to belong to the net space N bar p , bar q (M) and to the Lebesgue space L bar p 0,1 2 with mixed metric, where 1 bar p infty , 0 bar q leq infty , bar p (p 1,p 2) , bar q (q 1,q 2) , M is the set of all rectangles in mathbb R 2 . We proved the Hardy Littlewood theorem for multiple Fourier Haar series.

math.GR math.DS

Generating infinite monoids of cellular automata. For a group G and a set A , let text End (A G) be the monoid of all cellular automata over A G , and let text Aut (A G) be its group of units. By establishing a characterisation of surjunctuve groups in terms of the monoid text End (A G) , we prove that the rank of text End (A G) (i.e. the smallest cardinality of a generating set) is equal to the rank of text Aut (A G) plus the relative rank of text Aut (A G) in text End (A G) , and that the latter is infinite when G has an infinite decreasing chain of normal subgroups of finite index, condition which is satisfied, for example, for any infinite residually finite group. Moreover, when A V is a vector space over a field mathbb F , we study the monoid text End mathbb F (V G) of all linear cellular automata over V G and its group of units text Aut mathbb F (V G) . We show that if G is an indicable group and V is finite dimensional, then text End mathbb F (V G) is not finitely generated however, for any finitely generated indicable group G , the group text Aut mathbb F ( mathbb F G) is finitely generated if and only if mathbb F is finite.

q-bio.QM

Incorporating domain growth into hybrid methods for reaction diffusion systems. Reaction diffusion mechanism are a robust paradigm that can be used to represent many biological and physical phenomena over multiple spatial scales. Applications include intracellular dynamics, the migration of cells and the patterns formed by vegetation in semi arid landscapes. Moreover, domain growth is an important process for embryonic growth and wound healing. There are many numerical modelling frameworks capable of simulating such systems on growing domains, however each of these may be well suited to different spatial scales and particle numbers. Recently, spatially extended hybrid methods on static domains have been produced in order to bridge the gap between these different modelling paradigms in order to represent multiscale phenomena. However, such methods have not been developed with domain growth in mind. In this paper, we develop three hybrid methods on growing domains, extending three of the prominent static domain hybrid methods. We also provide detailed algorithms to allow others to employ them. We demonstrate that the methods are able to accurately model three representative reaction diffusion systems accurately and without bias.

cs.CL

AdapterFusion Non Destructive Task Composition for Transfer Learning. Sequential fine tuning and multi task learning are methods aiming to incorporate knowledge from multiple tasks however, they suffer from catastrophic forgetting and difficulties in dataset balancing. To address these shortcomings, we propose AdapterFusion, a new two stage learning algorithm that leverages knowledge from multiple tasks. First, in the knowledge extraction stage we learn task specific parameters called adapters, that encapsulate the task specific information. We then combine the adapters in a separate knowledge composition step. We show that by separating the two stages, i.e., knowledge extraction and knowledge composition, the classifier can effectively exploit the representations learned from multiple tasks in a non destructive manner. We empirically evaluate AdapterFusion on 16 diverse NLU tasks, and find that it effectively combines various types of knowledge at different layers of the model. We show that our approach outperforms traditional strategies such as full fine tuning as well as multi task learning. Our code and adapters are available at AdapterHub.ml.

math.AT cs.CG

Lifespan Functors and Natural Dualities in Persistent Homology. We introduce lifespan functors, which are endofunctors on the category of persistence modules that filter out intervals from barcodes according to their boundedness properties. They can be used to classify injective and projective objects in the category of barcodes and the category of pointwise finite dimensional persistence modules. They also naturally appear in duality results for absolute and relative versions of persistent (co)homology, generalizing previous results in terms of barcodes. Due to their functoriality, we can apply these results to morphisms in persistent homology that are induced by morphisms between filtrations. This lays the groundwork for the efficient computation of barcodes for images, kernels, and cokernels of such morphisms.

math.NT

On Dirichlet biquadratic fields. We study the 4 rank of the ideal class group of K n mathbb Q ( sqrt n , sqrt n ) . Our main result is that for a positive proportion of the squarefree integers n we have that the 4 rank of text Cl (K n) equals omega 3(n) 1 , where omega 3(n) is the number of prime divisors of n that are 3 modulo 4 .

cs.NE cs.LG

Boosting Throughput and Efficiency of Hardware Spiking Neural Accelerators using Time Compression Supporting Multiple Spike Codes. Spiking neural networks (SNNs) are the third generation of neural networks and can explore both rate and temporal coding for energy efficient event driven computation. However, the decision accuracy of existing SNN designs is contingent upon processing a large number of spikes over a long period. Nevertheless, the switching power of SNN hardware accelerators is proportional to the number of spikes processed while the length of spike trains limits throughput and static power efficiency. This paper presents the first study on developing temporal compression to significantly boost throughput and reduce energy dissipation of digital hardware SNN accelerators while being applicable to multiple spike codes. The proposed compression architectures consist of low cost input spike compression units, novel input and output weighted spiking neurons, and reconfigurable time constant scaling to support large and flexible time compression ratios. Our compression architectures can be transparently applied to any given pre designed SNNs employing either rate or temporal codes while incurring minimal modification of the neural models, learning algorithms, and hardware design. Using spiking speech and image recognition datasets, we demonstrate the feasibility of supporting large time compression ratios of up to 16x, delivering up to 15.93x, 13.88x, and 86.21x improvements in throughput, energy dissipation, the tradeoffs between hardware area, runtime, energy, and classification accuracy, respectively based on different spike codes on a Xilinx Zynq 7000 FPGA. These results are achieved while incurring little extra hardware overhead.

math.OC cs.LG

Follow the bisector a simple method for multi objective optimization. This study presents a novel Equiangular Direction Method (EDM) to solve a multi objective optimization problem. We consider optimization problems, where multiple differentiable losses have to be minimized. The presented method computes descent direction in every iteration to guarantee equal relative decrease of objective functions. This descent direction is based on the normalized gradients of the individual losses. Therefore, it is appropriate to solve multi objective optimization problems with multi scale losses. We test the proposed method on the imbalanced classification problem and multi task learning problem, where standard datasets are used. EDM is compared with other methods to solve these problems.

math.CO

Combinatorially refine a Zagier Stanley result on products of permutations. In this paper, we enumerate the pairs of permutations that are long cycles and whose product has a given cycle type. Our main result is a simple relation concerning the desired numbers for a few related cycle types. The relation refines a formula of the number of pairs of long cycles whose product has k cycles independently obtained by Zagier and Stanley relying on group characters, and was previously obtained by F ' e ray and Vassilieva by counting some colored permutations first and then relying on some algebraic computations in the ring of symmetric functions. Our approach here is simpler and combinatorial.

cs.LG

A Generalization of Transformer Networks to Graphs. We propose a generalization of transformer neural network architecture for arbitrary graphs. The original transformer was designed for Natural Language Processing (NLP), which operates on fully connected graphs representing all connections between the words in a sequence. Such architecture does not leverage the graph connectivity inductive bias, and can perform poorly when the graph topology is important and has not been encoded into the node features. We introduce a graph transformer with four new properties compared to the standard model. First, the attention mechanism is a function of the neighborhood connectivity for each node in the graph. Second, the positional encoding is represented by the Laplacian eigenvectors, which naturally generalize the sinusoidal positional encodings often used in NLP. Third, the layer normalization is replaced by a batch normalization layer, which provides faster training and better generalization performance. Finally, the architecture is extended to edge feature representation, which can be critical to tasks s.a. chemistry (bond type) or link prediction (entity relationship in knowledge graphs). Numerical experiments on a graph benchmark demonstrate the performance of the proposed graph transformer architecture. This work closes the gap between the original transformer, which was designed for the limited case of line graphs, and graph neural networks, that can work with arbitrary graphs. As our architecture is simple and generic, we believe it can be used as a black box for future applications that wish to consider transformer and graphs.

cs.CV cs.RO

Virtual Testbed for Monocular Visual Navigation of Small Unmanned Aircraft Systems. Monocular visual navigation methods have seen significant advances in the last decade, recently producing several real time solutions for autonomously navigating small unmanned aircraft systems without relying on GPS. This is critical for military operations which may involve environments where GPS signals are degraded or denied. However, testing and comparing visual navigation algorithms remains a challenge since visual data is expensive to gather. Conducting flight tests in a virtual environment is an attractive solution prior to committing to outdoor testing. This work presents a virtual testbed for conducting simulated flight tests over real world terrain and analyzing the real time performance of visual navigation algorithms at 31 Hz. This tool was created to ultimately find a visual odometry algorithm appropriate for further GPS denied navigation research on fixed wing aircraft, even though all of the algorithms were designed for other modalities. This testbed was used to evaluate three current state of the art, open source monocular visual odometry algorithms on a fixed wing platform Direct Sparse Odometry, Semi Direct Visual Odometry, and ORB SLAM2 (with loop closures disabled).

math.KT math.OA

Baum Connes and the Fourier Mukai transform. The Baum Connes map for finitely generated free abelian groups is a K theoretic analogue of the Fourier Mukai transform from algebraic geometry. We describe this K theoretic transform in the language of topological correspondences, and compute its action on K theory (of tori) described geometrically in terms of Baum Douglas cocycles, showing that the Fourier Mukai transform maps the class of a subtorus to the class of a suitably defined dual torus. We deduce the Fourier Mukai inversion formula. We use these results to give a purely geometric description of the Baum Connes assembly map for free abelian groups.

cs.CL cs.CV

Multimodal Pretraining Unmasked A Meta Analysis and a Unified Framework of Vision and Language BERTs. Large scale pretraining and task specific fine tuning is now the standard methodology for many tasks in computer vision and natural language processing. Recently, a multitude of methods have been proposed for pretraining vision and language BERTs to tackle challenges at the intersection of these two key areas of AI. These models can be categorised into either single stream or dual stream encoders. We study the differences between these two categories, and show how they can be unified under a single theoretical framework. We then conduct controlled experiments to discern the empirical differences between five V L BERTs. Our experiments show that training data and hyperparameters are responsible for most of the differences between the reported results, but they also reveal that the embedding layer plays a crucial role in these massive models.

math.OC

Pre Shape Calculus Foundations and Application to Mesh Quality Optimization. Deformations of the computational mesh arising from optimization routines usually lead to decrease of mesh quality or even destruction of the mesh. We propose a theoretical framework using pre shapes to generalize classical shape optimization and calculus. We define pre shape derivatives and derive according structure and calculus theorems. In particular, tangential directions are featured in pre shape derivatives, in contrast to classical shape derivatives featuring only normal directions. Techniques from classical shape optimization and calculus are shown to carry over to this framework. An optimization problem class for mesh quality is introduced, which is solvable by use of pre shape derivatives. This class allows for simultaneous optimization of classical shape objectives and mesh quality without deteriorating the classical shape optimization solution. The new techniques are implemented and numerically tested for 2D and 3D.

math.AG

Free dihedral actions on Abelian varieties. We give a simple construction for hyperelliptic varieties defined as the quotient of a complex torus by the action of a dihedral group that contains no translations and fixes no points. This generalizes a construction given by Catanese and Demleitner for D 4 in dimension three.

cs.LG cs.CR

Are Adversarial Examples Created Equal? A Learnable Weighted Minimax Risk for Robustness under Non uniform Attacks. Adversarial Training is proved to be an efficient method to defend against adversarial examples, being one of the few defenses that withstand strong attacks. However, traditional defense mechanisms assume a uniform attack over the examples according to the underlying data distribution, which is apparently unrealistic as the attacker could choose to focus on more vulnerable examples. We present a weighted minimax risk optimization that defends against non uniform attacks, achieving robustness against adversarial examples under perturbed test data distributions. Our modified risk considers importance weights of different adversarial examples and focuses adaptively on harder examples that are wrongly classified or at higher risk of being classified incorrectly. The designed risk allows the training process to learn a strong defense through optimizing the importance weights. The experiments show that our model significantly improves state of the art adversarial accuracy under non uniform attacks without a significant drop under uniform attacks.

cs.DM math.CO q-bio.PE

Combining Networks using Cherry Picking Sequences. Phylogenetic networks are important for the study of evolution. The number of methods to find such networks is increasing, but most such methods can only reconstruct small networks. To find bigger networks, one can attempt to combine small networks. In this paper, we study the sc Network Hybridization problem, a problem of combining networks into another network with low complexity. We characterize this complexity via a restricted problem, sc Tree child Network Hybridization , and we present an FPT algorithm to efficiently solve this restricted problem.

cs.IR cs.CL cs.LG

Knowledge Distillation in Document Retrieval. Complex deep learning models now achieve state of the art performance for many document retrieval tasks. The best models process the query or claim jointly with the document. However for fast scalable search it is desirable to have document embeddings which are independent of the claim. In this paper we show that knowledge distillation can be used to encourage a model that generates claim independent document encodings to mimic the behavior of a more complex model which generates claim dependent encodings. We explore this approach in document retrieval for a fact extraction and verification task. We show that by using the soft labels from a complex cross attention teacher model, the performance of claim independent student LSTM or CNN models is improved across all the ranking metrics. The student models we use are 12x faster in runtime and 20x smaller in number of parameters than the teacher

math.GR

An algorithm for finding minimal generating sets of finite groups. In this article, we study connections between components of the Cayley graph mathrm Cay (G,A) , where A is an arbitrary subset of a group G , and cosets of the subgroup of G generated by A . In particular, we show how to construct generating sets of G if mathrm Cay (G,A) has finitely many components. Furthermore, we provide an algorithm for finding minimal generating sets of finite groups using their Cayley graphs.

cs.CV cs.LG

Long Tailed Recognition Using Class Balanced Experts. Deep learning enables impressive performance in image recognition using large scale artificially balanced datasets. However, real world datasets exhibit highly class imbalanced distributions, yielding two main challenges relative imbalance amongst the classes and data scarcity for mediumshot or fewshot classes. In this work, we address the problem of long tailed recognition wherein the training set is highly imbalanced and the test set is kept balanced. Differently from existing paradigms relying on data resampling, cost sensitive learning, online hard example mining, loss objective reshaping, and or memory based modeling, we propose an ensemble of class balanced experts that combines the strength of diverse classifiers. Our ensemble of class balanced experts reaches results close to state of the art and an extended ensemble establishes a new state of the art on two benchmarks for long tailed recognition. We conduct extensive experiments to analyse the performance of the ensembles, and discover that in modern large scale datasets, relative imbalance is a harder problem than data scarcity. The training and evaluation code is available at https github.com ssfootball04 class balanced experts.

cs.SE

Agile Risk Management for Multi Cloud Software Development. Industry in all sectors is experiencing a profound digital transformation that puts software at the core of their businesses. In order to react to continuously changing user requirements and dynamic markets, companies need to build robust workflows that allow them to increase their agility in order to remain competitive. This increasingly rapid transformation, especially in domains like IoT or Cloud computing, poses significant challenges to guarantee high quality software, since dynamism and agile short term planning reduce the ability to detect and manage risks. In this paper, we describe the main challenges related to managing risk in agile software development, building on the experience of more than 20 agile coaches operating continuously for 15 years with hundreds of teams in industries in all sectors. We also propose a framework to manage risks that considers those challenges and supports collaboration, agility, and continuous development. An implementation of that framework is then described in a tool that handles risks and mitigation actions associated with the development of multi cloud applications. The methodology and the tool have been validated by a team of evaluators that were asked to consider its use in developing an urban smart mobility service and an airline flight scheduling system.

math.OC

An abstract lagrangian framework for computing shape derivatives. In this paper we study an abstract framework for computing shape derivatives of functionals subject to PDE constraints. We revisit the Lagrangian approach using the implicit function theorem in an abstract setting tailored for applications to shape optimization. This abstract framework yields practical formulae to compute the derivative of a shape functional, the material derivative of the state, and the adjoint state. Furthermore, it allows to gain insight on the duality between the material derivative of the state and the adjoint state. We show several applications of our main result to the computation of distributed shape derivatives for problems involving linear elliptic, nonlinear elliptic, parabolic PDEs and distributions. We compare our approach with other techniques for computing shape derivatives including the material derivative method and the averaged adjoint method.

math.AG

Moduli Problem of Hitchin Pairs over Deligne Mumford Stack. We define the moduli problem of Hitchin pairs over Deligne Mumford Stack and prove this moduli problem is represented by a separated and locally finitely presented algebraic space, which is considered as the moduli space of Hitchin pairs over Deligne Mumford stack.