Computer Science

Shaped by human movement, place connectivity is quantified by the strength of spatial interactions among locations. For decades, spatial scientists have researched place connectivity, applications, and metrics. The growing popularity of social media provides a new data stream where spatial social interaction measures are largely devoid of privacy issues, easily assessable, and harmonized. In this study, we introduced a global multi-scale place connectivity index (PCI) based on spatial interactions among places revealed by geotagged tweets as a spatiotemporal-continuous and easy-to-implement measurement. The multi-scale PCI, demonstrated at the US county level, exhibits a strong positive association with SafeGraph population movement records (10 percent penetration in the US population) and Facebook's social connectedness index (SCI), a popular connectivity index based on social networks. We found that PCI has a strong boundary effect and that it generally follows the distance decay, although this force is weaker in more urbanized counties with a denser population. Our investigation further suggests that PCI has great potential in addressing real-world problems that require place connectivity knowledge, exemplified with two applications: 1) modeling the spatial spread of COVID-19 during the early stage of the pandemic and 2) modeling hurricane evacuation destination choice. The methodological and contextual knowledge of PCI, together with the launched visualization platform and open-sourced PCI datasets at various geographic levels, are expected to support research fields requiring knowledge in human spatial interactions.

Wristwatches have been a common fashion accessory addition for several people. However, the concept of using a seven-segment digital display or sometimes, even an analog indicator hasn't changed for a number of years. This project aims to test and design a binary clock, also referred to as 32, 16, 8, 4, 2, 1 clock or even 8, 4, 2, 1 clock (due to their display configuration), that could change this everlasting display for watches. Specifically, digital logic and design engineers would find interest in this topic due to the sophistication involved in reading-out the time. This project will do so using by showing each decimal digit of sexagesimal time as a binary value. This design will be primarily functioning on logic gates and would involve the use of several basic components that include, but are not limited to, integrated circuits (or ICs), Light-emitting diodes (LEDs), and resistors.

Non-intrusive load monitoring (NILM) is a key cost-effective technology for monitoring power consumption and contributing to several challenges encountered when transiting to an efficient, sustainable, and competitive energy efficiency environment. This paper proposes a smart NILM system based on a novel local power histogramming (LPH) descriptor, in which appliance power signals are transformed into 2D space and short histograms are extracted to represent each device. Specifically, short local histograms are drawn to represent individual appliance consumption signatures and robustly extract appliance-level data from the aggregated power signal. Furthermore, an improved k-nearest neighbors (IKNN) algorithm is presented to reduce the learning computation time and improve the classification performance. This results in highly improving the discrimination ability between appliances belonging to distinct categories. A deep evaluation of the proposed LPH-IKNN based solution is investigated under different data sets, in which the proposed scheme leads to promising performance. An accuracy of up to 99.65% and 98.51% has been achieved on GREEND and UK-DALE data sets, respectively. While an accuracy of more than 96% has been attained on both WHITED and PLAID data sets. This proves the validity of using 2D descriptors to accurately identify appliances and create new perspectives for the NILM problem.

We study several variants of the problem of moving a convex polytope K , with n edges, in three dimensions through a flat rectangular (and sometimes more general) window. Specifically: ??We study variants where the motion is restricted to translations only, discuss situations where such a motion can be reduced to sliding (translation in a fixed direction), and present efficient algorithms for those variants, which run in time close to O( n 8/3 ) . ??We consider the case of a `gate' (an unbounded window with two parallel infinite edges), and show that K can pass through such a window, by any collision-free rigid motion, if and only if it can slide through it. ??We consider arbitrary compact convex windows, and show that if K can pass through such a window W (by any motion) then K can slide through a gate of width equal to the diameter of W . ??We study the case of a circular window W , and show that, for the regular tetrahedron K of edge length 1 , there are two thresholds 1> δ 1 ??.901388> δ 2 ??.895611 , such that (a) K can slide through W if the diameter d of W is ?? , (b) K cannot slide through W but can pass through it by a purely translational motion when δ 1 ?�d<1 , (c) K cannot pass through W by a purely translational motion but can do it when rotations are allowed when δ 2 ?�d< δ 1 , and (d) K cannot pass through W at all when d< δ 2 . ??Finally, we explore the general setup, where we want to plan a general motion (with all six degrees of freedom) for K through a rectangular window W , and present an efficient algorithm for this problem, with running time close to O( n 4 ) .

Within the context of stochastic probing with commitment, we consider the online stochastic matching problem for bipartite graphs where edges adjacent to an online node must be probed to determine if they exist, based on known edge probabilities. If a probed edge exists, it must be used in the matching (if possible). In addition to improving upon existing stochastic bipartite matching results, our results can also be seen as extensions to multi-item prophet inequalities. We study this matching problem for given constraints on the allowable sequences of probes adjacent to an online node. Our setting generalizes the patience (or time-out) constraint which limits the number of probes that can be made to edges. The generality of our setting leads to some modelling and computational efficiency issues that are not encountered in previous works. We establish new competitive bounds all of which generalize the standard non-stochastic setting when edges do not need to be probed (i.e., exist with certainty). Specifically, we establish the following competitive ratio results for a general formulation of edge constraints, arbitrary edge weights, and arbitrary edge probabilities: (1) A tight 1 2 ratio when the stochastic graph is generated from a known stochastic type graph where the ?(i ) th online node is drawn independently from a known distribution D ?(i) and ? is chosen adversarially. We refer to this setting as the known i.d. stochastic matching problem with adversarial arrivals. (2) A 1??/e ratio when the stochastic graph is generated from a known stochastic type graph where the ?(i ) th online node is drawn independently from a known distribution D ?(i) and ? is a random permutation. This is referred to as the known i.d. stochastic matching problem with random order arrivals.

Investigative Journalism (IJ, in short) is staple of modern, democratic societies. IJ often necessitates working with large, dynamic sets of heterogeneous, schema-less data sources, which can be structured, semi-structured, or textual, limiting the applicability of classical data integration approaches. In prior work, we have developed ConnectionLens, a system capable of integrating such sources into a single heterogeneous graph, leveraging Information Extraction (IE) techniques; users can then query the graph by means of keywords, and explore query results and their neighborhood using an interactive GUI. Our keyword search problem is complicated by the graph heterogeneity, and by the lack of a result score function that would allow to prune some of the search space. In this work, we describe an actual IJ application studying conflicts of interest in the biomedical domain, and we show how ConnectionLens supports it. Then, we present novel techniques addressing the scalability challenges raised by this application: one allows to reduce the significant IE costs while building the graph, while the other is a novel, parallel, in-memory keyword search engine, which achieves orders of magnitude speed-up over our previous engine. Our experimental study on the real-world IJ application data confirms the benefits of our contributions.

As artificial neural networks (ANNs) continue to make strides in wide-ranging and diverse fields of technology, the search for more efficient hardware implementations beyond conventional electronics is gaining traction. In particular, optical implementations potentially offer extraordinary gains in terms of speed and reduced energy consumption due to intrinsic parallelism of free-space optics. At the same time, a physical nonlinearity, a crucial ingredient of an ANN, is not easy to realize in free-space optics, which restricts the potential of this platform. This problem is further exacerbated by the need to perform the nonlinear activation also in parallel for each data point to preserve the benefit of linear free-space optics. Here, we present a free-space optical ANN with diffraction-based linear weight summation and nonlinear activation enabled by the saturable absorption of thermal atoms. We demonstrate, via both simulation and experiment, image classification of handwritten digits using only a single layer and observed 6-percent improvement in classification accuracy due to the optical nonlinearity compared to a linear model. Our platform preserves the massive parallelism of free-space optics even with physical nonlinearity, and thus opens the way for novel designs and wider deployment of optical ANNs.

The two-dimensional (2D) numerical approaches for vocal tract (VT) modelling can afford a better balance between the low computational cost and accurate rendering of acoustic wave propagation. However, they require a high spatio-temporal resolution in the numerical scheme for a precise estimation of acoustic formants at the simulation run-time expense. We have recently proposed a new VT acoustic modelling technique, known as the 2.5D Finite-Difference Time-Domain (2.5D FDTD), which extends the existing 2D FDTD approach by adding tube depth to its acoustic wave solver. In this work, first, the simulated acoustic outputs of our new model are shown to be comparable with the 2D FDTD and a realistic 3D FEM VT model at a low spatio-temporal resolution. Next, a radiation model is developed by including a circular baffle around the VT as head geometry. The transfer functions of the radiation model are analyzed using five different vocal tract shapes for vowel sounds /a/, /e/, /i/, /o/ and /u/.

Point cloud upsampling is vital for the quality of the mesh in three-dimensional reconstruction. Recent research on point cloud upsampling has achieved great success due to the development of deep learning. However, the existing methods regard point cloud upsampling of different scale factors as independent tasks. Thus, the methods need to train a specific model for each scale factor, which is both inefficient and impractical for storage and computation in real applications. To address this limitation, in this work, we propose a novel method called ``Meta-PU" to firstly support point cloud upsampling of arbitrary scale factors with a single model. In the Meta-PU method, besides the backbone network consisting of residual graph convolution (RGC) blocks, a meta-subnetwork is learned to adjust the weights of the RGC blocks dynamically, and a farthest sampling block is adopted to sample different numbers of points. Together, these two blocks enable our Meta-PU to continuously upsample the point cloud with arbitrary scale factors by using only a single model. In addition, the experiments reveal that training on multiple scales simultaneously is beneficial to each other. Thus, Meta-PU even outperforms the existing methods trained for a specific scale factor only.

Most computer science conferences rely on paper bidding to assign reviewers to papers. Although paper bidding enables high-quality assignments in days of unprecedented submission numbers, it also opens the door for dishonest reviewers to adversarially influence paper reviewing assignments. Anecdotal evidence suggests that some reviewers bid on papers by "friends" or colluding authors, even though these papers are outside their area of expertise, and recommend them for acceptance without considering the merit of the work. In this paper, we study the efficacy of such bid manipulation attacks and find that, indeed, they can jeopardize the integrity of the review process. We develop a novel approach for paper bidding and assignment that is much more robust against such attacks. We show empirically that our approach provides robustness even when dishonest reviewers collude, have full knowledge of the assignment system's internal workings, and have access to the system's inputs. In addition to being more robust, the quality of our paper review assignments is comparable to that of current, non-robust assignment approaches.

Technological progress has led to powerful computers and communication technologies that penetrate nowadays all areas of science, industry and our private lives. As a consequence, all these areas are generating digital traces of data amounting to big data resources. This opens unprecedented opportunities but also challenges toward the analysis, management, interpretation and utilization of these data. Fortunately, recent breakthroughs in deep learning algorithms complement now machine learning and statistics methods for an efficient analysis of such data. Furthermore, advances in text mining and natural language processing, e.g., word-embedding methods, enable also the processing of large amounts of text data from diverse sources as governmental reports, blog entries in social media or clinical health records of patients. In this paper, we present a perspective on the role of artificial intelligence in these developments and discuss also potential problems we are facing in a digital society.

Many real-world scenarios involve teams of agents that have to coordinate their actions to reach a shared goal. We focus on the setting in which a team of agents faces an opponent in a zero-sum, imperfect-information game. Team members can coordinate their strategies before the beginning of the game, but are unable to communicate during the playing phase of the game. This is the case, for example, in Bridge, collusion in poker, and collusion in bidding. In this setting, model-free RL methods are oftentimes unable to capture coordination because agents' policies are executed in a decentralized fashion. Our first contribution is a game-theoretic centralized training regimen to effectively perform trajectory sampling so as to foster team coordination. When team members can observe each other actions, we show that this approach provably yields equilibrium strategies. Then, we introduce a signaling-based framework to represent team coordinated strategies given a buffer of past experiences. Each team member's policy is parametrized as a neural network whose output is conditioned on a suitable exogenous signal, drawn from a learned probability distribution. By combining these two elements, we empirically show convergence to coordinated equilibria in cases where previous state-of-the-art multi-agent RL algorithms did not.

The overhead of the kernel storage path accounts for half of the access latency for new NVMe storage devices. We explore using BPF to reduce this overhead, by injecting user-defined functions deep in the kernel's I/O processing stack. When issuing a series of dependent I/O requests, this approach can increase IOPS by over 2.5 ? and cut latency by half, by bypassing kernel layers and avoiding user-kernel boundary crossings. However, we must avoid losing important properties when bypassing the file system and block layer such as the safety guarantees of the file system and translation between physical blocks addresses and file offsets. We sketch potential solutions to these problems, inspired by exokernel file systems from the late 90s, whose time, we believe, has finally come!

Most evolutionary algorithms have parameters, which allow a great flexibility in controlling their behavior and adapting them to new problems. To achieve the best performance, it is often needed to control some of the parameters during optimization, which gave rise to various parameter control methods. In recent works, however, similar advantages have been shown, and even proven, for sampling parameter values from certain, often heavy-tailed, fixed distributions. This produced a family of algorithms currently known as "fast evolution strategies" and "fast genetic algorithms". However, only little is known so far about the influence of these distributions on the performance of evolutionary algorithms, and about the relationships between (dynamic) parameter control and (static) parameter sampling. We contribute to the body of knowledge by presenting, for the first time, an algorithm that computes the optimal static distributions, which describe the mutation operator used in the well-known simple (1+λ) evolutionary algorithm on a classic benchmark problem OneMax. We show that, for large enough population sizes, such optimal distributions may be surprisingly complicated and counter-intuitive. We investigate certain properties of these distributions, and also evaluate the performance regrets of the (1+λ) evolutionary algorithm using commonly used mutation distributions.

Research has shown that trust is an essential aspect of human-computer interaction directly determining the degree to which the person is willing to use a system. An automatic prediction of the level of trust that a user has on a certain system could be used to attempt to correct potential distrust by having the system take relevant actions like, for example, apologizing or explaining its decisions. In this work, we explore the feasibility of automatically detecting the level of trust that a user has on a virtual assistant (VA) based on their speech. We developed a novel protocol for collecting speech data from subjects induced to have different degrees of trust in the skills of a VA. The protocol consists of an interactive session where the subject is asked to respond to a series of factual questions with the help of a virtual assistant. In order to induce subjects to either trust or distrust the VA's skills, they are first informed that the VA was previously rated by other users as being either good or bad; subsequently, the VA answers the subjects' questions consistently to its alleged abilities. All interactions are speech-based, with subjects and VAs communicating verbally, which allows the recording of speech produced under different trust conditions. Using this protocol, we collected a speech corpus in Argentine Spanish. We show clear evidence that the protocol effectively succeeded in influencing subjects into the desired mental state of either trusting or distrusting the agent's skills, and present results of a perceptual study of the degree of trust performed by expert listeners. Finally, we found that the subject's speech can be used to detect which type of VA they were using, which could be considered a proxy for the user's trust toward the VA's abilities, with an accuracy up to 76%, compared to a random baseline of 50%.

Similarity caching systems have recently attracted the attention of the scientific community, as they can be profitably used in many application contexts, like multimedia retrieval, advertising, object recognition, recommender systems and online content-match applications. In such systems, a user request for an object o , which is not in the cache, can be (partially) satisfied by a similar stored object o ', at the cost of a loss of user utility. In this paper we make a first step into the novel area of similarity caching networks, where requests can be forwarded along a path of caches to get the best efficiency-accuracy tradeoff. The offline problem of content placement can be easily shown to be NP-hard, while different polynomial algorithms can be devised to approach the optimal solution in discrete cases. As the content space grows large, we propose a continuous problem formulation whose solution exhibits a simple structure in a class of tree topologies. We verify our findings using synthetic and realistic request traces.

Data variety is one of the most important features of Big Data. Data variety is the result of aggregating data from multiple sources and uneven distribution of data. This feature of Big Data causes high variation in the consumption of processing resources such as CPU consumption. This issue has been overlooked in previous works. To overcome the mentioned problem, in the present work, we used Dynamic Voltage and Frequency Scaling (DVFS) to reduce the energy consumption of computation. To this goal, we consider two types of deadlines as our constraint. Before applying the DVFS technique to computer nodes, we estimate the processing time and the frequency needed to meet the deadline. In the evaluation phase, we have used a set of data sets and applications. The experimental results show that our proposed approach surpasses the other scenarios in processing real datasets. Based on the experimental results in this paper, DV-DVFS can achieve up to 15% improvement in energy consumption.

Food waste is a major challenge for the present world. It is the precursor to several socioeconomic problems that are plaguing the modern society. To counter the same and to, simultaneously, stand by the undernourished, surplus food redistribution has surfaced as a viable solution. Information and Communications Technology (ICT)-mediated food redistribution is a highly scalable approach and it percolates into the masses far better. Even if ICT is not brought into the picture, the presence of food surplus redistribution in developing countries like India is scarce and is limited to only a few of the major cities. The discussion of a surplus food redistribution framework under strategic settings is a less discussed topic around the globe. This paper aims at addressing a surplus food redistribution framework under strategic settings, thereby facilitating a smoother exchange of surplus food in the smart cities of developing countries, and beyond. As ICT is seamlessly available in smart cities, the paper aims to focus the framework in these cities. However, this can be extended beyond the smart cities to places with greater human involvement.

This paper presents a simple unsupervised visual representation learning method with a pretext task of discriminating all images in a dataset using a parametric, instance-level classifier. The overall framework is a replica of a supervised classification model, where semantic classes (e.g., dog, bird, and ship) are replaced by instance IDs. However, scaling up the classification task from thousands of semantic labels to millions of instance labels brings specific challenges including 1) the large-scale softmax computation; 2) the slow convergence due to the infrequent visiting of instance samples; and 3) the massive number of negative classes that can be noisy. This work presents several novel techniques to handle these difficulties. First, we introduce a hybrid parallel training framework to make large-scale training feasible. Second, we present a raw-feature initialization mechanism for classification weights, which we assume offers a contrastive prior for instance discrimination and can clearly speed up converge in our experiments. Finally, we propose to smooth the labels of a few hardest classes to avoid optimizing over very similar negative pairs. While being conceptually simple, our framework achieves competitive or superior performance compared to state-of-the-art unsupervised approaches, i.e., SimCLR, MoCoV2, and PIC under ImageNet linear evaluation protocol and on several downstream visual tasks, verifying that full instance classification is a strong pretraining technique for many semantic visual tasks.

Modern and emerging architectures demand increasingly complex compiler analyses and transformations. As the emphasis on compiler infrastructure moves beyond support for peephole optimizations and the extraction of instruction-level parallelism, they should support custom tools designed to meet these demands with higher-level analysis-powered abstractions of wider program scope. This paper introduces NOELLE, a robust open-source domain-independent compilation layer built upon LLVM providing this support. NOELLE is modular and demand-driven, making it easy-to-extend and adaptable to custom-tool-specific needs without unduly wasting compile time and memory. This paper shows the power of NOELLE by presenting a diverse set of ten custom tools built upon it, with a 33.2% to 99.2% reduction in code size (LoC) compared to their counterparts without NOELLE.

Embeddings of graphs into distributions of trees that preserve distances in expectation are a cornerstone of many optimization algorithms. Unfortunately, online or dynamic algorithms which use these embeddings seem inherently randomized and ill-suited against adaptive adversaries. In this paper we provide a new tree embedding which addresses these issues by deterministically embedding a graph into a single tree containing O(logn) copies of each vertex while preserving the connectivity structure of every subgraph and O( log 2 n) -approximating the cost of every subgraph. Using this embedding we obtain several new algorithmic results: We reduce an open question of Alon et al. [SODA 2004] -- the existence of a deterministic poly-log-competitive algorithm for online group Steiner tree on a general graph -- to its tree case. We give a poly-log-competitive deterministic algorithm for a closely related problem -- online partial group Steiner tree -- which, roughly, is a bicriteria version of online group Steiner tree. Lastly, we give the first poly-log approximations for demand-robust Steiner forest, group Steiner tree and group Steiner forest.

Quantile regression is a technique to analyse the effects of a set of independent variables on the entire distribution of a continuous response variable. Quantile regression presents a complete picture of the effects on the location, scale, and shape of the dependent variable at all points, not just at the mean. This research focuses on two challenges for the analysis of citation counts by quantile regression: discontinuity and substantial mass points at lower counts, such as zero, one, two, and three. A Bayesian two-part hurdle quantile regression model was proposed by King and Song (2019) as a suitable candidate for modeling count data with a substantial mass point at zero. Their model allows the zeros and non-zeros to be modeled independently but simultaneously. It uses quantile regression for modeling the nonzero data and logistic regression for modeling the probability of zeros versus nonzeros. Nevertheless, the current paper shows that substantial mass points also at one, two, and three for citation counts will nearly certainly affect the estimation of parameters in the quantile regression part of the model in a similar manner to the mass point at zero. We update the King and Song model by shifting the hurdle point from zero to three, past the main mass points. The new model delivers more accurate quantile regression for moderately to highly cited articles, and enables estimates of the extent to which factors influence the chances that an article will be low cited. To illustrate the advantage and potential of this method, it is applied separately to both simulated citation counts and also seven Scopus fields with collaboration, title length, and journal internationality as independent variables.

We present a general framework for modelling and verifying epistemic properties over parameterized multi-agent systems that communicate by truthful public announcements. In our framework, the number of agents or the amount of certain resources are parameterized (i.e. not known a priori), and the corresponding verification problem asks whether a given epistemic property is true regardless of the instantiation of the parameters. For example, in a muddy children puzzle, one could ask whether each child will eventually find out whether (s)he is muddy, regardless of the number of children. Our framework is regular model checking (RMC)-based, wherein synchronous finite-state automata (equivalently, monadic second-order logic over words) are used to specify the systems. We propose an extension of public announcement logic as specification language. Of special interests is the addition of the so-called iterated public announcement operators, which are crucial for reasoning about knowledge in parameterized systems. Although the operators make the model checking problem undecidable, we show that this becomes decidable when an appropriate "disappearance relation" is given. Further, we show how Angluin's L*-algorithm for learning finite automata can be applied to find a disappearance relation, which is guaranteed to terminate if it is regular. We have implemented the algorithm and apply this to such examples as the Muddy Children Puzzle, the Russian Card Problem, and Large Number Challenge.

Gamification has been used to motivate and engage participants in software engineering education and practice activities. There is a significant demand for empirical studies for the understanding of the impacts and efficacy of gamification. However, the lack of standard procedures and models for the evaluation of gamification is a challenge for the design, comparison, and report of results related to the assessment of gamification approaches and its effects. The goal of this study is to identify models and strategies for the evaluation of gamification reported in the literature. To achieve this goal, we conducted a systematic mapping study to investigate strategies for the evaluation of gamification in the context of software engineering. We selected 100 primary studies on gamification in software engineering (from 2011 to 2020). We categorized the studies regarding the presence of evaluation procedures or models for the evaluation of gamification, the purpose of the evaluation, the criteria used, the type of data, instruments, and procedures for data analysis. Our results show that 64 studies report procedures for the evaluation of gamification. However, only three studies actually propose evaluation models for gamification. We observed that the evaluation of gamification focuses on two aspects: the evaluation of the gamification strategy itself, related to the user experience and perceptions; and the evaluation of the outcomes and effects of gamification on its users and context. The most recurring criteria for the evaluation are 'engagement', 'motivation', 'satisfaction', and 'performance'. Finally, the evaluation of gamification requires a mix of subjective and objective inputs, and qualitative and quantitative data analysis approaches. Depending of the focus of the evaluation (the strategy or the outcomes), there is a predominance of a type of data and analysis.

No polynomial-time algorithm is known to test whether a sparse polynomial G divides another sparse polynomial F . While computing the quotient Q=F quo G can be done in polynomial time with respect to the sparsities of F, G and Q, this is not yet sufficient to get a polynomial-time divisibility test in general. Indeed, the sparsity of the quotient Q can be exponentially larger than the ones of F and G. In the favorable case where the sparsity #Q of the quotient is polynomial, the best known algorithm to compute Q has a non-linear factor #G#Q in the complexity, which is not optimal. In this work, we are interested in the two aspects of this problem. First, we propose a new randomized algorithm that computes the quotient of two sparse polynomials when the division is exact. Its complexity is quasi-linear in the sparsities of F, G and Q. Our approach relies on sparse interpolation and it works over any finite field or the ring of integers. Then, as a step toward faster divisibility testing, we provide a new polynomial-time algorithm when the divisor has a specific shape. More precisely, we reduce the problem to finding a polynomial S such that QS is sparse and testing divisibility by S can be done in polynomial time. We identify some structure patterns in the divisor G for which we can efficiently compute such a polynomial~S.

In order to achieve high accuracy for machine learning (ML) applications, it is essential to employ models with a large number of parameters. Certain applications, such as Automatic Speech Recognition (ASR), however, require real-time interactions with users, hence compelling the model to have as low latency as possible. Deploying large scale ML applications thus necessitates model quantization and compression, especially when running ML models on resource constrained devices. For example, by forcing some of the model weight values into zero, it is possible to apply zero-weight compression, which reduces both the model size and model reading time from the memory. In the literature, such methods are referred to as sparse pruning. The fundamental questions are when and which weights should be forced to zero, i.e. be pruned. In this work, we propose a compressed sensing based pruning (CSP) approach to effectively address those questions. By reformulating sparse pruning as a sparsity inducing and compression-error reduction dual problem, we introduce the classic compressed sensing process into the ML model training process. Using ASR task as an example, we show that CSP consistently outperforms existing approaches in the literature.

In this paper, we propose a novel multi-color balance method for reducing color distortions caused by lighting effects. The proposed method allows us to adjust three target-colors chosen by a user in an input image so that each target color is the same as the corresponding destination (benchmark) one. In contrast, white balancing is a typical technique for reducing the color distortions, however, they cannot remove lighting effects on colors other than white. In an experiment, the proposed method is demonstrated to be able to remove lighting effects on selected three colors, and is compared with existing white balance adjustments.

Enabling fully autonomous robots capable of navigating and exploring large-scale, unknown and complex environments has been at the core of robotics research for several decades. A key requirement in autonomous exploration is building accurate and consistent maps of the unknown environment that can be used for reliable navigation. Loop closure detection, the ability to assert that a robot has returned to a previously visited location, is crucial for consistent mapping as it reduces the drift caused by error accumulation in the estimated robot trajectory. Moreover, in multi-robot systems, loop closures enable merging local maps obtained by a team of robots into a consistent global map of the environment. In this paper, we present a degeneracy-aware and drift-resilient loop closing method to improve place recognition and resolve 3D location ambiguities for simultaneous localization and mapping (SLAM) in GPS-denied, large-scale and perceptually-degraded environments. More specifically, we focus on SLAM in subterranean environments (e.g., lava tubes, caves, and mines) that represent examples of complex and ambiguous environments where current methods have inadequate performance.

Efficient optimization of resources is paramount to success in many problems faced today. In the field of operational research the efficient scheduling of employees; packing of vans; routing of vehicles; logistics of airlines and transport of materials can be the difference between emission reduction or excess, profits or losses and feasibility or unworkable solutions. The video game Factorio, by Wube Software, has a myriad of problems which are analogous to such real-world problems, and is a useful simulator for developing solutions for these problems. In this paper we define the logistic transport belt problem and define mathematical integer programming model of it. We developed an interface to allow optimizers in any programming language to interact with Factorio, and we provide an initial benchmark of logistic transport belt problems. We present results for Simulated Annealing, quick Genetic Programming and Evolutionary Reinforcement Learning, three different meta-heuristic techniques to optimize this novel problem.

We consider reliable and secure communication over intersymbol interference wiretap channels (ISI-WTCs). In particular, we first examine the setup where the source at the input of an ISI-WTC is unconstrained and then, based on a general achievability result for arbitrary wiretap channels, we derive an achievable secure information rate for this ISI-WTC. Afterwards, we examine the setup where the source at the input of an ISI-WTC is constrained to be a finite-state machine source (FSMS) of a certain order and structure. Optimizing the parameters of this FSMS toward maximizing the secure information rate is a computationally intractable problem in general, and so, toward finding a local maximum, we propose an iterative algorithm that at every iteration replaces the secure information rate function by a suitable surrogate function whose maximum can be found efficiently. Although the secure information rates achieved in the unconstrained setup are expected to be larger than the secure information rates achieved in the constrained setup, the latter setup has the advantage of leading to efficient algorithms for estimating achievable secure rates and also has the benefit of being the basis of efficient encoding and decoding schemes.

We develop a computational framework that leverages the features of sophisticated software tools and numerics to tackle some of the pressing issues in the realm of earth sciences. The algorithms to handle the physics of multiphase flow, concomitant geomechanics all the way to the surface of the earth and the complex geometries of field cases with surfaces of discontinuity are stacked on top of each other in a modular fashion which allows for easy use to the end user. The current focus of the framework is to provide the user with tools for assessing seismic risks associated with energy technologies as well as for use in generating forward simulations in inversion analysis from data obtained using GPS and InSAR. In this work, we focus on one critical aspect in the development of the framework: the use of computational geometry in a two-grid method for unstructured tetrahedral meshes

Hate speech is increasingly prevalent online, and its negative outcomes include increased prejudice, extremism, and even offline hate crime. Automatic detection of online hate speech can help us to better understand these impacts. However, while the field has recently progressed through advances in natural language processing, challenges still remain. In particular, most existing approaches for hate speech detection focus on a single social media platform in isolation. This limits both the use of these models and their validity, as the nature of language varies from platform to platform. Here we propose a new cross-platform approach to detect hate speech which leverages multiple datasets and classification models from different platforms and trains a superlearner that can combine existing and novel training data to improve detection and increase model applicability. We demonstrate how this approach outperforms existing models, and achieves good performance when tested on messages from novel social media platforms not included in the original training data.

We present systematic and efficient solutions for both observability enhancement and root-cause diagnosis of post-silicon System-on-Chips (SoCs) validation with diverse usage scenarios. We model specification of interacting flows in typical applications for message selection. Our method for message selection optimizes flow specification coverage and trace buffer utilization. We define the diagnosis problem as identifying buggy traces as outliers and bug-free traces as inliers/normal behaviors, for which we use unsupervised learning algorithms for outlier detection. Instead of direct application of machine learning algorithms over trace data using the signals as raw features, we use feature engineering to transform raw features into more sophisticated features using domain specific operations. The engineered features are highly relevant to the diagnosis task and are generic to be applied across any hardware designs. We present debugging and root cause analysis of subtle post-silicon bugs in industry-scale OpenSPARC T2 SoC. We achieve a trace buffer utilization of 98.96\% with a flow specification coverage of 94.3\% (average). Our diagnosis method was able to diagnose up to 66.7\% more bugs and took up to 847 ? less diagnosis time as compared to the manual debugging with a diagnosis precision of 0.769.

Protecting privileged communications and data from disclosure is paramount for legal teams. Legal advice, such as attorney-client communications or litigation strategy are typically exempt from disclosure in litigations or regulatory events and are vital to the attorney-client relationship. To protect this information from disclosure, companies and outside counsel often review vast amounts of documents to determine those that contain privileged material. This process is extremely costly and time consuming. As data volumes increase, legal counsel normally employs methods to reduce the number of documents requiring review while balancing the need to ensure the protection of privileged information. Keyword searching is relied upon as a method to target privileged information and reduce document review populations. Keyword searches are effective at casting a wide net but often return overly inclusive results - most of which do not contain privileged information. To overcome the weaknesses of keyword searching, legal teams increasingly are using machine learning techniques to target privileged information. In these studies, classic text classification techniques are applied to build classification models to identify privileged documents. In this paper, the authors propose a different method by applying machine learning / convolutional neural network techniques (CNN) to identify privileged documents. Our proposed method combines keyword searching with CNN. For each keyword term, a CNN model is created using the context of the occurrences of the keyword. In addition, a method was proposed to select reliable privileged (positive) training keyword occurrences from labeled positive training documents. Extensive experiments were conducted, and the results show that the proposed methods can significantly reduce false positives while still capturing most of the true positives.

Automatic differentiation is a set of techniques to efficiently and accurately compute the derivative of a function represented by a computer program. Existing C++ libraries for automatic differentiation (e.g. Adept, Stan Math Library), however, exhibit large memory consumptions and runtime performance issues. This paper introduces FastAD, a new C++ template library for automatic differentiation, that overcomes all of these challenges in existing libraries by using vectorization, simpler memory management using a fully expression-template-based design, and other compile-time optimizations to remove some run-time overhead. Benchmarks show that FastAD performs 2-10 times faster than Adept and 2-19 times faster than Stan across various test cases including a few real-world examples.

In this paper we introduce a typed, concurrent λ -calculus with references featuring explicit substitutions for variables and references. Alongside usual safety properties, we recover strong normalization. The proof is based on a reducibility technique and an original interactive property reminiscent of the Game Semantics approach.

Tensor algebra is widely used in many applications, such as scientific computing, machine learning, and data analytics. The tensors represented real-world data are usually large and sparse. There are tens of storage formats designed for sparse matrices and/or tensors and the performance of sparse tensor operations depends on a particular architecture and/or selected sparse format, which makes it challenging to implement and optimize every tensor operation of interest and transfer the code from one architecture to another. We propose a tensor algebra domain-specific language (DSL) and compiler infrastructure to automatically generate kernels for mixed sparse-dense tensor algebra operations, named COMET. The proposed DSL provides high-level programming abstractions that resemble the familiar Einstein notation to represent tensor algebra operations. The compiler performs code optimizations and transformations for efficient code generation while covering a wide range of tensor storage formats. COMET compiler also leverages data reordering to improve spatial or temporal locality for better performance. Our results show that the performance of automatically generated kernels outperforms the state-of-the-art sparse tensor algebra compiler, with up to 20.92x, 6.39x, and 13.9x performance improvement, for parallel SpMV, SpMM, and TTM over TACO, respectively.

The Stabbing Planes proof system was introduced to model the reasoning carried out in practical mixed integer programming solvers. As a proof system, it is powerful enough to simulate Cutting Planes and to refute the Tseitin formulas -- certain unsatisfiable systems of linear equations mod 2 -- which are canonical hard examples for many algebraic proof systems. In a recent (and surprising) result, Dadush and Tiwari showed that these short refutations of the Tseitin formulas could be translated into quasi-polynomial size and depth Cutting Planes proofs, refuting a long-standing conjecture. This translation raises several interesting questions. First, whether all Stabbing Planes proofs can be efficiently simulated by Cutting Planes. This would allow for the substantial analysis done on the Cutting Planes system to be lifted to practical mixed integer programming solvers. Second, whether the quasi-polynomial depth of these proofs is inherent to Cutting Planes. In this paper we make progress towards answering both of these questions. First, we show that any Stabbing Planes proof with bounded coefficients SP* can be translated into Cutting Planes. As a consequence of the known lower bounds for Cutting Planes, this establishes the first exponential lower bounds on SP*. Using this translation, we extend the result of Dadush and Tiwari to show that Cutting Planes has short refutations of any unsatisfiable system of linear equations over a finite field. Like the Cutting Planes proofs of Dadush and Tiwari, our refutations also incur a quasi-polynomial blow-up in depth, and we conjecture that this is inherent. As a step towards this conjecture, we develop a new geometric technique for proving lower bounds on the depth of Cutting Planes proofs. This allows us to establish the first lower bounds on the depth of Semantic Cutting Planes proofs of the Tseitin formulas.