arXiv subjects

Tamara Broderick

Publications and source records attributed to Tamara Broderick.

At least 91 records · Page 5Linked to original sources

arXiv2016

Edge-exchangeable graphs and sparsity

A known failing of many popular random graph models is that the Aldous-Hoover Theorem guarantees these graphs are dense with probability one; that is, the number of edges grows quadratically with the number of nodes. This behavior is considered unrealistic in observed graphs. We define a notion of edge exchangeability for random graphs in contrast to the established notion of infinite exchangeability for random graphs --- which has traditionally relied on exchangeability of nodes (rather than edges) in a graph. We show that, unlike node exchangeability, edge exchangeability encompasses models that are known to provide a projective sequence of random graphs that circumvent the Aldous-Hoover Theorem and exhibit sparsity, i.e., sub-quadratic growth of the number of edges with the number of nodes. We show how edge-exchangeability of graphs relates naturally to existing notions of exchangeability from clustering (a.k.a. partitions) and other familiar combinatorial structures.

math.ST↗

arXiv2016

Completely random measures for modeling power laws in sparse graphs

Network data appear in a number of applications, such as online social networks and biological networks, and there is growing interest in both developing models for networks as well as studying the properties of such data. Since individual network datasets continue to grow in size, it is necessary to develop models that accurately represent the real-life scaling properties of networks. One behavior of interest is having a power law in the degree distribution. However, other types of power laws that have been observed empirically and considered for applications such as clustering and feature allocation models have not been studied as frequently in models for graph data. In this paper, we enumerate desirable asymptotic behavior that may be of interest for modeling graph data, including sparsity and several types of power laws. We outline a general framework for graph generative models using completely random measures; by contrast to the pioneering work of Caron and Fox (2015), we consider instantiating more of the existing atoms of the random measure as the dataset size increases rather than adding new atoms to the measure. We see that these two models can be complementary; they respectively yield interpretations as (1) time passing among existing members of a network and (2) new individuals joining a network. We detail a particular instance of this framework and show simulated results that suggest this model exhibits some desirable asymptotic power-law behavior.

stat.ML↗

arXiv2015

Linear Response Methods for Accurate Covariance Estimates from Mean Field Variational Bayes

Mean field variational Bayes (MFVB) is a popular posterior approximation method due to its fast runtime on large-scale data sets. However, it is well known that a major failing of MFVB is that it underestimates the uncertainty of model variables (sometimes severely) and provides no information about model variable covariance. We generalize linear response methods from statistical physics to deliver accurate uncertainty estimates for model variables---both for individual variables and coherently across variables. We call our method linear response variational Bayes (LRVB). When the MFVB posterior approximation is in the exponential family, LRVB has a simple, analytic form, even for non-conjugate models. Indeed, we make no assumptions about the form of the true posterior. We demonstrate the accuracy and scalability of our method on a range of models for both simulated and real data.

stat.ML↗

arXiv2015

Robust Inference with Variational Bayes

In Bayesian analysis, the posterior follows from the data and a choice of a prior and a likelihood. One hopes that the posterior is robust to reasonable variation in the choice of prior and likelihood, since this choice is made by the modeler and is necessarily somewhat subjective. Despite the fundamental importance of the problem and a considerable body of literature, the tools of robust Bayes are not commonly used in practice. This is in large part due to the difficulty of calculating robustness measures from MCMC draws. Although methods for computing robustness measures from MCMC draws exist, they lack generality and often require additional coding or computation. In contrast to MCMC, variational Bayes (VB) techniques are readily amenable to robustness analysis. The derivative of a posterior expectation with respect to a prior or data perturbation is a measure of local robustness to the prior or likelihood. Because VB casts posterior inference as an optimization problem, its methodology is built on the ability to calculate derivatives of posterior quantities with respect to model parameters, even in very complex models. In the present work, we develop local prior robustness measures for mean-field variational Bayes(MFVB), a VB technique which imposes a particular factorization assumption on the variational posterior approximation. We start by outlining existing local prior measures of robustness. Next, we use these results to derive closed-form measures of the sensitivity of mean-field variational posterior approximation to prior specification. We demonstrate our method on a meta-analysis of randomized controlled interventions in access to microcredit in developing countries.

stat.ME↗

arXiv2015

A translation of "The characteristic function of a random phenomenon" by Bruno de Finetti

This article is a translation of Bruno de Finetti's paper "Funzione Caratteristica di un fenomeno aleatorio" which appeared in Atti del Congresso Internazionale dei Matematici, Bologna 3-10 Settembre 1928, Tomo VI, pp. 179-190, originally published by Nicola Zanichelli Editore S.p.A. The translation was made as close as possible to the original in form and style, except for apparent mistakes found in the original document, which were corrected and are mentioned as footnotes. Most of these were resolved by comparing against a longer version of this work by de Finetti, published shortly after this one under the same titlea. The interested reader is highly encouraged to consult this other version for a more detailed treatment of the topics covered here. Footnotes regarding the translation are labeled with letters to distinguish them from de Finetti's original footnotes.

math.ST↗

arXiv2015

Covariance Matrices and Influence Scores for Mean Field Variational Bayes

Mean field variational Bayes (MFVB) is a popular posterior approximation method due to its fast runtime on large-scale data sets. However, it is well known that a major failing of MFVB is that it underestimates the uncertainty of model variables (sometimes severely) and provides no information about model variable covariance. We develop a fast, general methodology for exponential families that augments MFVB to deliver accurate uncertainty estimates for model variables -- both for individual variables and coherently across variables. MFVB for exponential families defines a fixed-point equation in the means of the approximating posterior, and our approach yields a covariance estimate by perturbing this fixed point. Inspired by linear response theory, we call our method linear response variational Bayes (LRVB). We also show how LRVB can be used to quickly calculate a measure of the influence of individual data points on parameter point estimates. We demonstrate the accuracy and scalability of our method by learning Gaussian mixture models for both simulated and real data.

stat.ML↗

arXiv2014

Covariance Matrices for Mean Field Variational Bayes

Mean Field Variational Bayes (MFVB) is a popular posterior approximation method due to its fast runtime on large-scale data sets. However, it is well known that a major failing of MFVB is its (sometimes severe) underestimates of the uncertainty of model variables and lack of information about model variable covariance. We develop a fast, general methodology for exponential families that augments MFVB to deliver accurate uncertainty estimates for model variables -- both for individual variables and coherently across variables. MFVB for exponential families defines a fixed-point equation in the means of the approximating posterior, and our approach yields a covariance estimate by perturbing this fixed point. Inspired by linear response theory, we call our method linear response variational Bayes (LRVB). We demonstrate the accuracy of our method on simulated data sets.

stat.ML↗

arXiv2014

Variational Bayes for Merging Noisy Databases

Bayesian entity resolution merges together multiple, noisy databases and returns the minimal collection of unique individuals represented, together with their true, latent record values. Bayesian methods allow flexible generative models that share power across databases as well as principled quantification of uncertainty for queries of the final, resolved database. However, existing Bayesian methods for entity resolution use Markov monte Carlo method (MCMC) approximations and are too slow to run on modern databases containing millions or billions of records. Instead, we propose applying variational approximations to allow scalable Bayesian inference in these models. We derive a coordinate-ascent approximation for mean-field variational Bayes, qualitatively compare our algorithm to existing methods, note unique challenges for inference that arise from the expected distribution of cluster sizes in entity resolution, and discuss directions for future work in this domain.

stat.ME↗

arXiv2013

Streaming Variational Bayes

We present SDA-Bayes, a framework for (S)treaming, (D)istributed, (A)synchronous computation of a Bayesian posterior. The framework makes streaming updates to the estimated posterior according to a user-specified approximation batch primitive. We demonstrate the usefulness of our framework, with variational Bayes (VB) as the primitive, by fitting the latent Dirichlet allocation model to two large-scale document collections. We demonstrate the advantages of our algorithm over stochastic variational inference (SVI) by comparing the two after a single pass through a known amount of data---a case where SVI may be applied---and in the streaming setting, where SVI does not apply.

stat.ML↗

arXiv2013

Cluster and Feature Modeling from Combinatorial Stochastic Processes

One of the focal points of the modern literature on Bayesian nonparametrics has been the problem of clustering, or partitioning, where each data point is modeled as being associated with one and only one of some collection of groups called clusters or partition blocks. Underlying these Bayesian nonparametric models are a set of interrelated stochastic processes, most notably the Dirichlet process and the Chinese restaurant process. In this paper we provide a formal development of an analogous problem, called feature modeling, for associating data points with arbitrary nonnegative integer numbers of groups, now called features or topics. We review the existing combinatorial stochastic process representations for the clustering problem and develop analogous representations for the feature modeling problem. These representations include the beta process and the Indian buffet process as well as new representations that provide insight into the connections between these processes. We thereby bring the same level of completeness to the treatment of Bayesian nonparametric feature modeling that has previously been achieved for Bayesian nonparametric clustering.

math.ST↗

arXiv2013

Optimistic Concurrency Control for Distributed Unsupervised Learning

Research on distributed machine learning algorithms has focused primarily on one of two extremes - algorithms that obey strict concurrency constraints or algorithms that obey few or no such constraints. We consider an intermediate alternative in which algorithms optimistically assume that conflicts are unlikely and if conflicts do arise a conflict-resolution protocol is invoked. We view this "optimistic concurrency control" paradigm as particularly appropriate for large-scale machine learning algorithms, particularly in the unsupervised setting. We demonstrate our approach in three problem areas: clustering, feature learning and online facility location. We evaluate our methods via large-scale experiments in a cluster computing environment.

cs.LG↗

arXiv2013

Combinatorial clustering and the beta negative binomial process

We develop a Bayesian nonparametric approach to a general family of latent class problems in which individuals can belong simultaneously to multiple classes and where each class can be exhibited multiple times by an individual. We introduce a combinatorial stochastic process known as the negative binomial process (NBP) as an infinite-dimensional prior appropriate for such problems. We show that the NBP is conjugate to the beta process, and we characterize the posterior distribution under the beta-negative binomial process (BNBP) and hierarchical models based on the BNBP (the HBNBP). We study the asymptotic properties of the BNBP and develop a three-parameter extension of the BNBP that exhibits power-law behavior. We derive MCMC algorithms for posterior inference under the HBNBP, and we present experiments using these algorithms in the domains of image segmentation, object recognition, and document analysis.

stat.ME↗

arXiv2013

MAD-Bayes: MAP-based Asymptotic Derivations from Bayes

The classical mixture of Gaussians model is related to K-means via small-variance asymptotics: as the covariances of the Gaussians tend to zero, the negative log-likelihood of the mixture of Gaussians model approaches the K-means objective, and the EM algorithm approaches the K-means algorithm. Kulis & Jordan (2012) used this observation to obtain a novel K-means-like algorithm from a Gibbs sampler for the Dirichlet process (DP) mixture. We instead consider applying small-variance asymptotics directly to the posterior in Bayesian nonparametric models. This framework is independent of any specific Bayesian inference algorithm, and it has the major advantage that it generalizes immediately to a range of models beyond the DP mixture. To illustrate, we apply our framework to the feature learning setting, where the beta process and Indian buffet process provide an appropriate Bayesian nonparametric prior. We obtain a novel objective function that goes beyond clustering to learn (and penalize new) groupings for which we relax the mutual exclusivity and exhaustivity assumptions of clustering. We demonstrate several other algorithms, all of which are scalable and simple to implement. Empirical results demonstrate the benefits of the new framework.

stat.ML↗

arXiv2013

Real-time semiparametric regression

We develop algorithms for performing semiparametric regression analysis in real time, with data processed as it is collected and made immediately available via modern telecommunications technologies. Our definition of semiparametric regression is quite broad and includes, as special cases, generalized linear mixed models, generalized additive models, geostatistical models, wavelet nonparametric regression models and their various combinations. Fast updating of regression fits is achieved by couching semiparametric regression into a Bayesian hierarchical model or, equivalently, graphical model framework and employing online mean field variational ideas. An internet site attached to this article, realtime-semiparametric-regression.net, illustrates the methodology for continually arriving stock market, real estate and airline data. Flexible real-time analyses, based on increasingly ubiquitous streaming data sources stand to benefit.

stat.ME↗

arXiv2013

Feature allocations, probability functions, and paintboxes

The problem of inferring a clustering of a data set has been the subject of much research in Bayesian analysis, and there currently exists a solid mathematical foundation for Bayesian approaches to clustering. In particular, the class of probability distributions over partitions of a data set has been characterized in a number of ways, including via exchangeable partition probability functions (EPPFs) and the Kingman paintbox. Here, we develop a generalization of the clustering problem, called feature allocation, where we allow each data point to belong to an arbitrary, non-negative integer number of groups, now called features or topics. We define and study an "exchangeable feature probability function" (EFPF)---analogous to the EPPF in the clustering setting---for certain types of feature models. Moreover, we introduce a "feature paintbox" characterization---analogous to the Kingman paintbox for clustering---of the class of exchangeable feature models. We provide a further characterization of the subclass of feature allocations that have EFPF representations.

math.PR↗

arXiv2012

Combining Spatial and Telemetric Features for Learning Animal Movement Models

We introduce a new graphical model for tracking radio-tagged animals and learning their movement patterns. The model provides a principled way to combine radio telemetry data with an arbitrary set of userdefined, spatial features. We describe an efficient stochastic gradient algorithm for fitting model parameters to data and demonstrate its effectiveness via asymptotic analysis and synthetic experiments. We also apply our model to real datasets, and show that it outperforms the most popular radio telemetry software package used in ecology. We conclude that integration of different data sources under a single statistical framework, coupled with appropriate parameter and state estimation procedures, produces both accurate location estimates and an interpretable statistical model of animal movement.

cs.LG↗

arXiv2011

Rapid, Machine-Learned Resource Allocation: Application to High-redshift GRB Follow-up

As the number of observed Gamma-Ray Bursts (GRBs) continues to grow, follow-up resources need to be used more efficiently in order to maximize science output from limited telescope time. As such, it is becoming increasingly important to rapidly identify bursts of interest as soon as possible after the event, before the afterglows fade beyond detectability. Studying the most distant (highest redshift) events, for instance, remains a primary goal for many in the field. Here we present our Random forest Automated Triage Estimator for GRB redshifts (RATE GRB-z) for rapid identification of high-redshift candidates using early-time metrics from the three telescopes onboard Swift. While the basic RATE methodology is generalizable to a number of resource allocation problems, here we demonstrate its utility for telescope-constrained follow-up efforts with the primary goal to identify and study high-z GRBs. For each new GRB, RATE GRB-z provides a recommendation - based on the available telescope time - of whether the event warrants additional follow-up resources. We train RATE GRB-z using a set consisting of 135 Swift bursts with known redshifts, only 18 of which are z > 4. Cross-validated performance metrics on this training data suggest that ~56% of high-z bursts can be captured from following up the top 20% of the ranked candidates, and ~84% of high-z bursts are identified after following up the top ~40% of candidates. We further use the method to rank 200+ Swift bursts with unknown redshifts according to their likelihood of being high-z.

astro-ph.IM↗

arXiv2011

Beta processes, stick-breaking, and power laws

The beta-Bernoulli process provides a Bayesian nonparametric prior for models involving collections of binary-valued features. A draw from the beta process yields an infinite collection of probabilities in the unit interval, and a draw from the Bernoulli process turns these into binary-valued features. Recent work has provided stick-breaking representations for the beta process analogous to the well-known stick-breaking representation for the Dirichlet process. We derive one such stick-breaking representation directly from the characterization of the beta process as a completely random measure. This approach motivates a three-parameter generalization of the beta process, and we study the power laws that can be obtained from this generalized beta process. We present a posterior inference algorithm for the beta-Bernoulli process that exploits the stick-breaking representation, and we present experimental results for a discrete factor-analysis model.

stat.ME↗