Search

Bayesian data analysis is about more than just computing a posteriordistribu- tion, and Bayesian visualization is about more than trace plots ofMarkov chains. Practical Bayesian data analysis, like all data analysis, is aniterative process of model building, inference, model checking and evaluation...

The present paper concerns large covariance matrix estimation via compositeminimization under the assumption of low rank plus sparse structure. In thisapproach, the low rank plus sparse decomposition of the covariance matrix isrecovered by least squares minimization under nuclear norm plus $l_1$...

We present a novel preconditioning technique for proximal optimizationmethods that relies on graph algorithms to construct effective preconditioners.Such combinatorial preconditioners arise from partitioning the graph intoforests. We prove that certain decompositions lead to a theoretically...

Modern stochastic optimization methods often rely on uniform sampling whichis agnostic to the underlying characteristics of the data. This might degradethe convergence by yielding estimates that suffer from a high variance. Apossible remedy is to employ non-uniform importance sampling techniques...

The sparse inverse covariance estimation problem is commonly solved using an$\ell_{1}$-regularized Gaussian maximum likelihood estimator known as"graphical lasso", but its computational cost becomes prohibitive for largedata sets. A recent line of results showed--under mild assumptions--that...

We propose a stepsize adaptation scheme for stochastic gradient descent. Itoperates directly with the loss function and rescales the gradient in order tomake fixed predicted progress on the loss. We demonstrate its capabilities bystrongly improving the performance of Adam and Momentum optimizers...

Training modern deep learning models requires large amounts of computation,often provided by GPUs. Scaling computation from one GPU to many can enablemuch faster training and research progress but entails two complications.First, the training library must support inter-GPU communication. Depending...

Shannon's mathematical theory of communication defines fundamental limits onhow much information can be transmitted between the different components of anyman-made or biological system. This paper is an informal but rigorousintroduction to the main ideas implicit in Shannon's theory. An...

The easiness at which adversarial instances can be generated in deep neuralnetworks raises some fundamental questions on their functioning and concerns ontheir use in critical systems. In this paper, we draw a connection betweenover-generalization and adversaries: a possible cause of adversaries...

Multi-output regression models must exploit dependencies between outputs tomaximise predictive performance. The application of Gaussian processes (GPs) tothis setting typically yields models that are computationally demanding andhave limited representational power. We present the Gaussian...

Despite their popularity, the practical performance of asynchronousstochastic gradient descent methods (ASGD) for solving large scale machinelearning problems are not as good as theoretical results indicate. We adopt andanalyze a synchronous K-step averaging stochastic gradient descent...

Superconductivity has been the focus of enormous research effort since itsdiscovery more than a century ago. Yet, some features of this unique phenomenonremain poorly understood; prime among these is the connection betweensuperconductivity and chemical/structural properties of materials. To...

There is increasing interest in learning algorithms that involve interactionbetween human and machine. Comparison-based queries are among the most naturalways to get feedback from humans. A challenge in designing comparison-basedinteractive learning algorithms is coping with noisy answers. The most...

Gravitational mass flows, such as avalanches, debris flows and rockfalls arecommon events in alpine regions with high impact on transport routes. Withinthe last few decades, hazard zone maps have been developed to systematicallyapproach this threat. These maps mark vulnerable zones in habitable...

The ridesharing economy is experiencing rapid growth and innovation.Companies such as Uber and Lyft are continuing to grow at a considerable pacewhile providing their platform as an organizing medium for ridesharingservices, increasing consumer utility as well as employing thousands inpart-time...

Rankings are ubiquitous in the online world today. As we have transitionedfrom finding books in libraries to ranking products, jobs, job applicants,opinions and potential romantic partners, there is a substantial precedent thatranking systems have a responsibility not only to their users but also to...

This article presents an overview of applications of logic programming,classifying them based on the abstractions and implementations of logiclanguages that support the applications. The three key abstractions are join,recursion, and constraint. Their essential implementations are for-loops...

Nowadays, the Internet is indispensable when it comes to informationdissemination. People rely on the Internet to inform themselves on current newsevents, as well as to verify facts. We, as a community, are quickly approachingan 'electronic information age' where the majority of information will...

Societies are complex systems which tend to polarize into sub-groups ofindividuals with dramatically opposite perspectives. This phenomenon isreflected -- and often amplified -- in online social networks where, however,humans are no more the only players, and co-exist alongside with social bots,i.e...

This is a survey of algorithmic problems in group theory, old and new,motivated by applications to cryptography.