We introduce DiaSet, a novel dataset of dialectical Arabic speech, manually transcribed and annotated for two specific downstream tasks: sentiment analysis and named entity recognition. The dataset encapsulates the Palestine dialect, predominantly spoken in Palestine, Israel, and Jordan. Our dataset incorporates authentic conversations between YouTube influencers and their respective guests. Furthermore, we have enriched the dataset with simulated conversations initiated by inviting participants from various locales within the said regions. The participants were encouraged to engage in dialogues with our interviewer. Overall, DiaSet consists of 644.8K tokens and 23.2K annotated instances. Uniform writing standards were upheld during the transcription process. Additionally, we established baseline models by leveraging some of the pre-existing Arabic BERT language models, showcasing the potential applications and efficiencies of our dataset. We make DiaSet publicly available for further research.

Time series classification problems are prevalent across various domains, often characterized by intra-series relationships within features, and inter-series relationships between the same features over time. Developing a generalized model capable of capturing these intricate properties poses a considerable challenge. In this paper, we introduce an innovative approach termed Gaussian Process Variational Information Bottleneck (GP-VIB). This model is designed as an end-to-end system, with a primary focus on acquiring a concise representation of the initial sequence. It aims to retain essential information vital for accurate classification. Through our experiments, we illustrate that the proposed GP-VIB model outperforms existing methods on renowned benchmark datasets.

We introduce a novel deep Mixture of Experts (MoE) architecture, termed MoEAtt, that integrates an attention mechanism as the routing gate, where the individual experts and the router are trained jointly. Furthermore, the training procedure is designed to achieve heterogeneity across the experts, which in turn yield a discriminative representation of the input space. We evaluate MoEAtt architecture on multiple datasets to demonstrate its versatility and applicability in various scenarios. We achieve state-of-the-art performance on certain datasets, showcasing the effectiveness of MoEAtt. Additionally, we discuss and showcase additional benefits and potential provided by the MoE architecture.

Variational inference provides a way to approximate probability densities. It does so by optimizing an upper or a lower bound on the likelihood of the observed data (the evidence). The classic variational inference approach suggests to maximize the Evidence Lower BOund (ELBO). Recent proposals suggest to optimize the variational Rényi bound (VR) and χ upper bound. However, these estimates are either biased or difficult to approximate, due to a high variance.In this paper we introduce a new upper bound (termed VRLU) which is based on the existing variational Rényi bound. In contrast to the existing VR bound, the Monte Carlo (MC) approximation of the VRLU bound is unbiased. Furthermore, we devise a (sandwiched) upper-lower bound variational inference method (termed VRS) to jointly optimize the upper and lower bounds. We present a set of experiments, designed to evaluate the new VRLU bound, and to compare the VRS method with the classic VAE and the VR methods over a set of digit recognition tasks. The experiments and results demonstrate the VRLU bound advantage, and the wide applicability of the VRS method.

Latent variables pose a challenge for accurate modelling, experimental design, and inference, since they may cause non-adjustable bias in the estimation of effects. While most of the research regarding latent variables revolves around accounting for their presence and learning how they interact with other variables in the experiment, their bare existence is assumed to be deduced based on domain expertise. In this work we focus on the discovery of such latent variables, utilizing statistical hypothesis testing methods and Bayesian Networks learning. Specifically, we present a novel method for detecting discrete latent factors which affect continuous observed outcomes, in mixed discrete/continuous observed data, and device a structure learning algorithm that adds the detected latent factors to a fully observed Bayesian Network. Finally, we demonstrate the utility of our method with a set of experiments, in both controlled and real-life settings, one of which is a prediction for the outcome of COVID-19 test results.

The weight decay regularization term is widely used during training to constrain expressivity, avoid overfitting, and improve generalization. Historically, this concept was borrowed from the SVM maximum margin principle and extended to multiclass deep networks. Carefully inspecting this principle reveals that it is not optimal for multi-class classification in general, and in particular when using deep neural networks. In this paper, we explain why this commonly used principle is not optimal and propose a new regularization scheme, called Pairwise Margin Maximization (PMM), which measures the minimal amount of displacement an instance should take until its predicted classification is switched. In deep neural networks, PMM can be implemented in the vector space before the network’s output layer, i.e., in the deep feature space, where we add an additional normalization term to avoid convergence to a trivial solution. We demonstrate empirically a substantial improvement when training a deep neural network with PMM compared to the standard regularization terms.

We study the decision problem of disjunctive linear arithmetic over the reals from the perspective of computational geometry. We show that traversing the linear arrangement induced by the formula's predicates, rather than the DPLL(T) method of traversing the Boolean space, may have an advantage when the number of variables is smaller than the number of predicates (as it is indeed the case in the standard SMT-Lib benchmarks). We then continue by showing a branching heuristic that is based on approximating T-implications, based on a geometric analysis. We achieve modest improvement in run time comparing to the commonly used heuristic used by competitive solvers.

The initial state of a design under verification has a major impact on the ability of stimuli generators to successfully generate the requested stimuli. For complexity reasons, most stimuli generators use sequential solutions without planning ahead. Therefore, in many cases, they fail to produce a consistent stimuli due to an inadequate selection of the initial state. We propose a new method, based on machine learning techniques, to improve generation success by learning the relationship between the initial state vector and generation success. We applied the proposed method in two different settings, with the objective of improving generation success and coverage in processor and system level generation. In both settings, the proposed method significantly reduced generation failures and enabled faster coverage

In this article we present novel learning methods for estimating the quality of results returned by a search engine in response to a query. Estimation is based on the agreement between the top results of the full query and the top results of its sub-queries. We demonstrate the usefulness of quality estimation for several applications, among them improvement of retrieval, detecting queries for which no relevant content exists in the document collection, and distributed information retrieval. Experiments on TREC data demonstrate the robustness and the effectiveness of our learning algorithms.

Random test generators are often used to create regression suites on-the-fly. Regression suites are commonly generated by choosing several specifications and generating a number of tests from each one, without reasoning which specification should be used and how many tests should be generated from each specification. This paper describes a technique for building high quality random regression suites. The proposed technique uses information about the probability of each test specification covering each coverage task. This probability is used, in turn, to determine which test specifications should be included in the regression suite and how many tests should be generated from each specification. Experimental results show that this practical technique can be used to improve the quality, and reduce the cost, of regression suites. Moreover, it enables better informed decisions regarding the size and distribution of the regression suites, and the risk involved.

Functional verification is widely acknowledged as the bottleneck in the hardware design cycle. This paper addresses one of the main challenges of simulation based verification (or dynamic verification), by providing a new approach for Coverage Directed Test Generation (CDG). This approach is based on Bayesian networks and computer learning techniques. It provides an efficient way for closing a feedback loop from the coverage domain back to a generator that produces new stimuli to the tested design. In this paper, we show how to apply Bayesian networks to the CDG problem. Applying Bayesian networks to the CDG framework has been tested in several experiments, exhibiting encouraging results and indicating that the suggested approach can be used to achieve CDG goals.

This paper exploits the fact that when GMM and SVM classifiers with roughly the same level of performance exhibit uncorrelated errors they can be combined to produce a better classifier. The gain accrues from combining the descriptive strength of GMM models with the discriminative power of SVM classifiers. This idea, first exploited in the context of speaker recognition [1, 2], is applied to speech recognition - specifically to a digit recognition task in a noisy environment - with significant gains in performance.

Proposes a classification scheme that incorporates statistical models and support vector machines. A hybrid system which appropriately combines the advantages of both the generative and discriminant model paradigms is described and experimentally evaluated on a text-independent speaker recognition task in matched and mismatched training and test conditions. Our results prove that the combination is beneficial in terms of performance and practical in terms of computation. We report relative improvements of up to 25% reduction in identification error rate compared to the baseline statistical model.

This paper proposes a classification scheme that combines statistical models and support vector machines. It exploits the fact (observed in [1]) that GMM and SVM classifiers with roughly the same level of performance produce uncorrelated errors. We describe a novel scheme which employs an SVM classifier as an “advisor ” to the GMM classifier in uncertain cases. The utility of the combined generative/discriminative approach is demonstrated on standard text-independent speaker verification and speaker identification tasks in matched and mismatched training and test conditions. Results indicate significant improvements in performance without much computational overhead. 1.

We propose a framework based on a parametric quadratic program(cid:173) ming (QP) technique to solve the support vector machine (SVM) training problem. This framework, can be specialized to obtain two SVM optimization methods. The first solves the fixed bias prob(cid:173) lem, while the second starts with an optimal solution for a fixed bias problem and adjusts the bias until the optimal value is found. The later method can be applied in conjunction with any other ex(cid:173) isting technique which obtains a fixed bias solution. Moreover, the second method can also be used independently to solve the com(cid:173) plete SVM training problem. A combination of these two methods is more flexible than each individual method and, among other things, produces an incremental algorithm which exactly solve the 1-Norm Soft Margin SVM optimization problem. Applying Selec(cid:173) tive Sampling techniques may further boost convergence.