MOTIVATION: Feature selection is a critical task in machine learning and statistics. However, existing feature selection methods either (i) rely on parametric methods such as linear or generalized linear models, (ii) lack theoretical false discovery control, or (iii) identify few true positives. RESULTS: We introduce a general feature selection method with finite-sample false discovery control based on applying integrated path stability selection (IPSS) to arbitrary feature importance scores. The method is nonparametric whenever the importance scores are nonparametric, and it estimates q-values, which are better suited to high-dimensional data than P-values. We focus on two special cases using importance scores from gradient boosting (IPSSGB) and random forests (IPSSRF). Extensive nonlinear simulations with RNA sequencing data show that both methods accurately control the false discovery rate and detect more true positives than existing methods. Both methods are also efficient, running in under 20 s when there are 500 samples and 5000 features. We apply IPSSGB and IPSSRF to detect microRNAs and genes related to cancer, finding that they yield better predictions with fewer features than existing approaches. AVAILABILITY AND IMPLEMENTATION: All code and data used in this work are available on GitHub (https://github.com/omelikechi/ipss_bioinformatics) and permanently archived on Zenodo (https://doi.org/10.5281/zenodo.15335289). A Python package for implementing IPSS is available on GitHub (https://github.com/omelikechi/ipss) and PyPI (https://pypi.org/project/ipss/). An R implementation of IPSS is also available on GitHub (https://github.com/omelikechi/ipssR).

In the current landscape of deep learning research, there is a predominant emphasis on achieving high predictive accuracy in supervised tasks involving large image and language datasets. However, a broader perspective reveals a multitude of overlooked metrics, tasks, and data types, such as uncertainty, active and continual learning, and scientific data, that demand attention. Bayesian deep learning (BDL) constitutes a promising avenue, offering advantages across these diverse settings. This paper posits that BDL can elevate the capabilities of deep learning. It revisits the strengths of BDL, acknowledges existing challenges, and highlights some exciting research avenues aimed at addressing these obstacles. Looking ahead, the discussion focuses on possible ways to combine large-scale foundation models with BDL to unlock their full potential.

Quantifying spatial and/or temporal associations in multivariate geolocated data of different types is achievable via spatial random effects in a Bayesian hierarchical model, but severe computational bottlenecks arise when spatial dependence is encoded as a latent Gaussian process (GP) in the increasingly common large scale data settings on which we focus. The scenario worsens in non-Gaussian models because the reduced analytical tractability leads to additional hurdles to computational efficiency. In this article, we introduce Bayesian models of spatially referenced data in which the likelihood or the latent process (or both) are not Gaussian. First, we exploit the advantages of spatial processes built via directed acyclic graphs, in which case the spatial nodes enter the Bayesian hierarchy and lead to posterior sampling via routine Markov chain Monte Carlo (MCMC) methods. Second, motivated by the possible inefficiencies of popular gradient-based sampling approaches in the multivariate contexts on which we focus, we introduce the simplified manifold preconditioner adaptation (SiMPA) algorithm which uses second order information about the target but avoids expensive matrix operations. We demostrate the performance and efficiency improvements of our methods relative to alternatives in extensive synthetic and real world remote sensing and community ecology applications with large scale data at up to hundreds of thousands of spatial locations and up to tens of outcomes. Software for the proposed methods is part of R package meshed, available on CRAN.

Bayesian mixture models are widely used for clustering of high-dimensional data with appropriate uncertainty quantification. However, as the dimension of the observations increases, posterior inference often tends to favor too many or too few clusters. This article explains this behavior by studying the random partition posterior in a non-standard setting with a fixed sample size and increasing data dimensionality. We provide conditions under which the finite sample posterior tends to either assign every observation to a different cluster or all observations to the same cluster as the dimension grows. Interestingly, the conditions do not depend on the choice of clustering prior, as long as all possible partitions of observations into clusters have positive prior probabilities, and hold irrespective of the true data-generating model. We then propose a class of latent mixtures for Bayesian clustering (Lamb) on a set of low-dimensional latent variables inducing a partition on the observed data. The model is amenable to scalable posterior inference and we show that it can avoid the pitfalls of high-dimensionality under mild assumptions. The proposed approach is shown to have good performance in simulation studies and an application to inferring cell types based on scRNAseq.

There is a rich literature on Bayesian methods for density estimation, which characterize the unknown density as a mixture of kernels. Such methods have advantages in terms of providing uncertainty quantification in estimation, while being adaptive to a rich variety of densities. However, relative to frequentist locally adaptive kernel methods, Bayesian approaches can be slow and unstable to implement in relying on Markov chain Monte Carlo algorithms. To maintain most of the strengths of Bayesian approaches without the computational disadvantages, we propose a class of nearest neighbor-Dirichlet mixtures. The approach starts by grouping the data into neighborhoods based on standard algorithms. Within each neighborhood, the density is characterized via a Bayesian parametric model, such as a Gaussian with unknown parameters. Assigning a Dirichlet prior to the weights on these local kernels, we obtain a pseudo-posterior for the weights and kernel parameters. A simple and embarrassingly parallel Monte Carlo algorithm is proposed to sample from the resulting pseudo-posterior for the unknown density. Desirable asymptotic properties are shown, and the methods are evaluated in simulation studies and applied to a motivating data set in the context of classification.

In multivariate data analysis, it is often important to estimate a graph characterizing dependence among $p$ variables. A popular strategy in Gaussian graphical models and latent Gaussian graphical models uses the non-zero entries in a $p\times p$ covariance or precision matrix, typically requiring restrictive modeling assumptions for accurate graph recovery. To improve model robustness, we instead focus on estimating the backbone of the dependence graph. We use a spanning tree likelihood, based on a minimalist graphical model that is purposely overly-simplified. Taking a Bayesian approach, we place a prior on the space of trees and quantify uncertainty in the graphical model. In both theory and experiments, we show that this model does not require the population graph to be a spanning tree or the covariance to satisfy assumptions beyond positive-definiteness. The model accurately recovers the backbone of the population graph at a rate competitive with existing approaches but with better robustness. We show combinatorial properties of the spanning tree, which may be of independent interest, and develop an efficient Gibbs sampler for Bayesian inference. Analyzing electroencephalography data using a hidden Markov model with each latent state modeled by a spanning tree, we show that results are much more interpretable compared with popular alternatives.

Mixed Membership Models (MMMs) are a popular family of latent structure models for complex multivariate data. Instead of forcing each subject to belong to a single cluster, MMMs incorporate a vector of subject-specific weights characterizing partial membership across clusters. With this flexibility come challenges in uniquely identifying, estimating, and interpreting the parameters. In this article, we propose a new class of Dimension-Grouped MMMs (Gro-M$^3$s) for multivariate categorical data, which improve parsimony and interpretability. In Gro-M$^3$s, observed variables are partitioned into groups such that the latent membership is constant for variables within a group but can differ across groups. Traditional latent class models are obtained when all variables are in one group, while traditional MMMs are obtained when each variable is in its own group. The new model corresponds to a novel decomposition of probability tensors. Theoretically, we derive transparent identifiability conditions for both the unknown grouping structure and model parameters in general settings. Methodologically, we propose a Bayesian approach for Dirichlet Gro-M$^3$s to inferring the variable grouping structure and estimating model parameters. Simulation results demonstrate good computational performance and empirically confirm the identifiability results. We illustrate the new methodology through applications to a functional disability survey dataset and a personality test dataset.

MOTIVATION: It has become routine in neuroscience studies to measure brain networks for different individuals using neuroimaging. These networks are typically expressed as adjacency matrices, with each cell containing a summary of connectivity between a pair of brain regions. There is an emerging statistical literature describing methods for the analysis of such multi-network data in which nodes are common across networks but the edges vary. However, there has been essentially no consideration of the important problem of outlier detection. In particular, for certain subjects, the neuroimaging data are so poor quality that the network cannot be reliably reconstructed. For such subjects, the resulting adjacency matrix may be mostly zero or exhibit a bizarre pattern not consistent with a functioning brain. These outlying networks may serve as influential points, contaminating subsequent statistical analyses. We propose a simple Outlier DetectIon for Networks (ODIN) method relying on an influence measure under a hierarchical generalized linear model for the adjacency matrices. An efficient computational algorithm is described, and ODIN is illustrated through simulations and an application to data from the UK Biobank. RESULTS: ODIN was successful in identifying moderate to extreme outliers. Removing such outliers can significantly change inferences in downstream applications. AVAILABILITY AND IMPLEMENTATION: ODIN has been implemented in both Python and R and these implementations along with other code are publicly available at github.com/pritamdey/ODIN-python and github.com/pritamdey/ODIN-r, respectively. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

High resolution geospatial data are challenging because standard geostatistical models based on Gaussian processes are known to not scale to large data sizes. While progress has been made towards methods that can be computed more efficiently, considerably less attention has been devoted to methods for large scale data that allow the description of complex relationships between several outcomes recorded at high resolutions by different sensors. Our Bayesian multivariate regression models based on spatial multivariate trees (SpamTrees) achieve scalability via conditional independence assumptions on latent random effects following a treed directed acyclic graph. Information-theoretic arguments and considerations on computational efficiency guide the construction of the tree and the related efficient sampling algorithms in imbalanced multivariate settings. In addition to simulated data examples, we illustrate SpamTrees using a large climate data set which combines satellite data with land-based station data. Software and source code are available on CRAN at https://CRAN.R-project.org/package=spamtree.

Transformation based methods have been an attractive approach in non-parametric inference for problems such as unconditioned and conditional density estimation due to their unique hierarchical structure that models the data as flexible transformation of a set of common latent variables. More recently, transformation based models have been used in variational inference (VI) to construct flexible implicit families of variational distributions. However, their use in both non-parametric inference and variational inference lacks theoretical justification. In the context of non-linear latent variable models (NL-LVM), we provide theoretical justification for the use of these models in non-parametric inference by showing that the support of the transformation induced prior in the space of densities is sufficiently large in the $L_1$ sense and show that for this class of priors the posterior concentrates at the optimal rate up to a logarithmic factor. Adopting the flexibility demonstrated in the non-parametric setting we use the NL-LVM to construct an implicit family of variational distributions, deemed as GP-IVI. We delineate sufficient conditions under which GP-IVI achieves optimal risk bounds and approximates the true posterior in the sense of the Kullback-Leibler divergence. To the best of our knowledge, this is the first work on providing theoretical guarantees for implicit variational inference.

Model-based clustering is widely used in a variety of application areas. However, fundamental concerns remain about robustness. In particular, results can be sensitive to the choice of kernel representing the within-cluster data density. Leveraging on properties of pairwise differences between data points, we propose a class of Bayesian distance clustering methods, which rely on modeling the likelihood of the pairwise distances in place of the original data. Although some information in the data is discarded, we gain substantial robustness to modeling assumptions. The proposed approach represents an appealing middle ground between distance- and model-based clustering, drawing advantages from each of these canonical approaches. We illustrate dramatic gains in the ability to infer clusters that are not well represented by the usual choices of kernel. A simulation study is included to assess performance relative to competitors, and we apply the approach to clustering of brain genome expression data.

Statistical methods relating tensor predictors to scalar outcomes in a regression model generally vectorize the tensor predictor and estimate the coefficients of its entries employing some form of regularization, use summaries of the tensor covariate, or use a low dimensional approximation of the coefficient tensor. However, low rank approximations of the coefficient tensor can suffer if the true rank is not small. We propose a tensor regression framework which assumes a soft version of the parallel factors (PARAFAC) approximation. In contrast to classic PARAFAC where each entry of the coefficient tensor is the sum of products of row-specific contributions across the tensor modes, the soft tensor regression (Softer) framework allows the row-specific contributions to vary around an overall mean. We follow a Bayesian approach to inference, and show that softening the PARAFAC increases model flexibility, leads to improved estimation of coefficient tensors, more accurate identification of important predictor entries, and more precise predictions, even for a low approximation rank. From a theoretical perspective, we show that employing Softer leads to a weakly consistent posterior distribution of the coefficient tensor, irrespective of the true or approximation tensor rank, a result that is not true when employing the classic PARAFAC for tensor regression. In the context of our motivating application, we adapt Softer to symmetric and semi-symmetric tensor predictors and analyze the relationship between brain network characteristics and human traits.

Many modern data sets require inference methods that can estimate the shared and individual-specific components of variability in collections of matrices that change over time. Promising methods have been developed to analyze these types of data in static cases, but only a few approaches are available for dynamic settings. To address this gap, we consider novel models and inference methods for pairs of matrices in which the columns correspond to multivariate observations at different time points. In order to characterize common and individual features, we propose a Bayesian dynamic factor modeling framework called Time Aligned Common and Individual Factor Analysis (TACIFA) that includes uncertainty in time alignment through an unknown warping function. We provide theoretical support for the proposed model, showing identifiability and posterior concentration. The structure enables efficient computation through a Hamiltonian Monte Carlo (HMC) algorithm. We show excellent performance in simulations, and illustrate the method through application to a social mimicry experiment.

We introduce a novel regularization approach for deep learning that incorporates and respects the underlying graphical structure of the neural network. Existing regularization methods often focus on penalizing weights in a global/uniform manner that ignores the connectivity structure of the neural network. We propose to use the Fiedler value of the neural network’s underlying graph as a tool for regularization. We provide theoretical support for this approach via spectral graph theory. We show several useful properties of the Fiedler value that make it suitable for regularization. We provide an approximate, variational approach for faster computation during training. We provide an alternative formulation of this framework in the form of a structurally weighted L1 penalty, thus linking our approach to sparsity induction. We performed experiments on datasets that compare Fiedler regularization with traditional regularization methods such as Dropout and weight decay. Results demonstrate the efficacy of Fiedler regularization.

MOTIVATION: Low-dimensional representations of high-dimensional data are routinely employed in biomedical research to visualize, interpret and communicate results from different pipelines. In this article, we propose a novel procedure to directly estimate t-SNE embeddings that are not driven by batch effects. Without correction, interesting structure in the data can be obscured by batch effects. The proposed algorithm can therefore significantly aid visualization of high-dimensional data. RESULTS: The proposed methods are based on linear algebra and constrained optimization, leading to efficient algorithms and fast computation in many high-dimensional settings. Results on artificial single-cell transcription profiling data show that the proposed procedure successfully removes multiple batch effects from t-SNE embeddings, while retaining fundamental information on cell types. When applied to single-cell gene expression data to investigate mouse medulloblastoma, the proposed method successfully removes batches related with mice identifiers and the date of the experiment, while preserving clusters of oligodendrocytes, astrocytes, and endothelial cells and microglia, which are expected to lie in the stroma within or adjacent to the tumours. AVAILABILITY AND IMPLEMENTATION: Source code implementing the proposed approach is available as an R package at https://github.com/emanuelealiverti/BC_tSNE, including a tutorial to reproduce the simulation studies. CONTACT: [email protected].

Closed surfaces provide a useful model for $3$-d shapes, with the data typically consisting of a cloud of points in $\mathbb{R}^3$. The existing literature on closed surface modeling focuses on frequentist point estimation methods that join surface patches along the edges, with surface patches created via Bézier surfaces or tensor products of B-splines. However, the resulting surfaces are not smooth along the edges and the geometric constraints required to join the surface patches lead to computational drawbacks. In this article, we develop a Bayesian model for closed surfaces based on tensor products of a cyclic basis resulting in infinitely smooth surface realizations. We impose sparsity on the control points through a double-shrinkage prior. Theoretical properties of the support of our proposed prior are studied and it is shown that the posterior achieves the optimal rate of convergence under reasonable assumptions on the prior. The proposed approach is illustrated with some examples.

Kernel mixture models are routinely used for density estimation. However, in multivariate settings, issues arise in efficiently approximating lower-dimensional structure in the data. For example, it is common to suppose that the density is concentrated near a lower-dimensional non-linear subspace or manifold. Typical kernels used to locally approximate such subspaces are inflexible, so that a large number of components are often needed. We propose a novel class of LOcally COnvex (LOCO) kernels that are flexible in adapting to nonlinear local structure. LOCO kernels are induced by introducing random knots within local neighborhoods, and generating data as a random convex combination of these knots with adaptive weights and an additive noise. For identifiability, we constrain all observations from a particular component to have the same mean. For Bayesian inference subject to this constraint, we develop a hybrid Gibbs sampler and optimization algorithm that incorporates a Lagrange multiplier within a splitting method. The resulting LOCO algorithm is shown to dramatically outperform typical Gaussian mixture models in challenging examples.

Motivation: Although there is a rich literature on methods for assessing the impact of functional predictors, the focus has been on approaches for dimension reduction that do not suit certain applications. Examples of standard approaches include functional linear models, functional principal components regression and cluster-based approaches, such as latent trajectory analysis. This article is motivated by applications in which the dynamics in a predictor, across times when the value is relatively extreme, are particularly informative about the response. For example, physicians are interested in relating the dynamics of blood pressure changes during surgery to post-surgery adverse outcomes, and it is thought that the dynamics are more important when blood pressure is significantly elevated or lowered. Results: We propose a novel class of extrema-weighted feature (XWF) extraction models. Key components in defining XWFs include the marginal density of the predictor, a function up-weighting values at extreme quantiles of this marginal, and functionals characterizing local dynamics. Algorithms are proposed for fitting of XWF-based regression and classification models, and are compared with current methods for functional predictors in simulations and a blood pressure during surgery application. XWFs find features of intraoperative blood pressure trajectories that are predictive of postoperative mortality. By their nature, most of these features cannot be found by previous methods. Availability and implementation: The R package 'xwf' is available at the CRAN repository: https://cran.r-project.org/package=xwf. Supplementary information: Supplementary data are available at Bioinformatics online.

There has been considerable interest in making Bayesian inference more scalable. In big data settings, most of the focus has been on reducing the computing time per iteration rather than reducing the number of iterations needed in Markov chain Monte Carlo (MCMC). This article considers data augmentation MCMC (DA-MCMC), a widely used technique. DA-MCMC samples tend to become highly autocorrelated in large samples, due to a mis-calibration problem in which conditional posterior distributions given augmented data are too concentrated. This makes it necessary to collect very long MCMC paths to obtain acceptably low MC error. To combat this inefficiency, we propose a family of calibrated data augmentation algorithms, which appropriately adjust the variance of conditional posterior distributions. A Metropolis-Hastings step is used to eliminate bias in the stationary distribution of the resulting sampler. Compared to existing alternatives, this approach can dramatically reduce MC error by reducing autocorrelation and increasing the effective number of DA-MCMC samples per unit of computing time. The approach is simple and applicable to a broad variety of existing data augmentation algorithms. We focus on three popular generalized linear models: probit, logistic and Poisson log-linear. Dramatic gains in computational efficiency are shown in applications.

Divide-and-conquer based methods for Bayesian inference provide a general approach for tractable posterior inference when the sample size is large. These methods divide the data into smaller subsets, sample from the posterior distribution of parameters in parallel on all the subsets, and combine posterior samples from all the subsets to approximate the full data posterior distribution. The smaller size of any subset compared to the full data implies that posterior sampling on any subset is computationally more efficient than sampling from the true posterior distribution. Since the combination step takes negligible time relative to sampling, posterior computations can be scaled to massive data by dividing the full data into sufficiently large number of data subsets. One such approach relies on the geometry of posterior distributions estimated across different subsets and combines them through their barycenter in a Wasserstein space of probability measures. We provide theoretical guarantees on the accuracy of approximation that are valid in many applications. We show that the geometric method approximates the full data posterior distribution better than its competitors across diverse simulations and reproduces known results when applied to a movie ratings database.

MOTIVATION: There is increasing interest in learning how human brain networks vary as a function of a continuous trait, but flexible and efficient procedures to accomplish this goal are limited. We develop a Bayesian semiparametric model, which combines low-rank factorizations and flexible Gaussian process priors to learn changes in the conditional expectation of a network-valued random variable across the values of a continuous predictor, while including subject-specific random effects. RESULTS: The formulation leads to a general framework for inference on changes in brain network structures across human traits, facilitating borrowing of information and coherently characterizing uncertainty. We provide an efficient Gibbs sampler for posterior computation along with simple procedures for inference, prediction and goodness-of-fit assessments. The model is applied to learn how human brain networks vary across individuals with different intelligence scores. Results provide interesting insights on the association between intelligence and brain connectivity, while demonstrating good predictive performance. AVAILABILITY AND IMPLEMENTATION: Source code implemented in R and data are available at https://github.com/wangronglu/BNRR. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

We propose a Bayesian approach to regression with a scalar response on vector and tensor covariates. Vectorization of the tensor prior to analysis fails to exploit the structure, often leading to poor estimation and predictive performance. We introduce a novel class of multiway shrinkage priors for tensor coefficients in the regression setting and present posterior consistency results under mild conditions. A computationally efficient Markov chain Monte Carlo algorithm is developed for posterior computation. Simulation studies illustrate substantial gains over existing tensor regression methods in terms of estimation and parameter inference. Our approach is further illustrated in a neuroimaging application.

We propose a novel approach to Bayesian analysis that is provably robust to outliers in the data and often has computational advantages over standard methods. Our technique is based on splitting the data into non-overlapping subgroups, evaluating the posterior distribution given each independent subgroup, and then combining the resulting measures. The main novelty of our approach is the proposed aggregation step, which is based on the evaluation of a median in the space of probability measures equipped with a suitable collection of distances that can be quickly and efficiently evaluated in practice. We present both theoretical and numerical evidence illustrating the improvements achieved by our method.

We utilize copulas to constitute a unified framework for constructing and optimizing variational proposals in hierarchical Bayesian models. For models with continuous and non-Gaussian hidden variables, we propose a semiparametric and automated variational Gaussian copula approach, in which the parametric Gaussian copula family is able to preserve multivariate posterior dependence, and the nonparametric transformations based on Bernstein polynomials provide ample flexibility in characterizing the univariate marginal posteriors.

It is standard to assume a low-dimensional structure in estimating a high-dimensional density. However, popular methods, such as probabilistic principal component analysis, scale poorly computationally. We introduce a novel empirical Bayes method that we term geometric density estimation (GEODE) and show that, with mild conditions and among all d-dimensional linear subspaces, the span of the d leading principal axes of the data maximizes the model posterior. With these axes pre-computed using fast singular value decomposition, GEODE easily scales to high dimensional problems while providing uncertainty characterization. The model is also capable of imputing missing data and dynamically deleting redundant dimensions. Finally, we generalize GEODE by mixing it across a dyadic clustering tree. Both simulation studies and real world data applications show superior performance of GEODE in terms of robustness and computational efficiency.

Ordinary least squares (OLS) is the default method for fitting linear models, but is not applicable for problems with dimensionality larger than the sample size. For these problems, we advocate the use of a generalized version of OLS motivated by ridge regression, and propose two novel three-step algorithms involving least squares fitting and hard thresholding. The algorithms are methodologically simple to understand intuitively, computationally easy to implement efficiently, and theoretically appealing for choosing models consistently. Numerical exercises comparing our methods with penalization-based approaches in simulations and data analyses illustrate the great potential of the proposed algorithms.

Fitting statistical models is computationally challenging when the sample size or the dimension of the dataset is huge. An attractive approach for down-scaling the problem size is to first partition the dataset into subsets and then fit using distributed algorithms. The dataset can be partitioned either horizontally (in the sample space) or vertically (in the feature space). While the majority of the literature focuses on sample space partitioning, feature space partitioning is more effective when p >> n. Existing methods for partitioning features, however, are either vulnerable to high correlations or inefficient in reducing the model dimension. In this paper, we solve these problems through a new embarrassingly parallel framework named DECO for distributed variable selection and parameter estimation. In DECO, variables are first partitioned and allocated to m distributed workers. The decorrelated subset data within each worker are then fitted via any algorithm designed for high-dimensional problems. We show that by incorporating the decorrelation step, DECO can achieve consistent variable selection and parameter estimation on each subset with (almost) no assumptions. In addition, the convergence rate is nearly minimax optimal for both sparse and weakly sparse models and does NOT depend on the partition number m. Extensive numerical experiments are provided to illustrate the performance of the new framework.

Nonparametric regression for large numbers of features ($p$) is an increasingly important problem. If the sample size $n$ is massive, a common strategy is to partition the feature space, and then separately apply simple models to each partition set. This is not ideal when $n$ is modest relative to $p$, and we propose an alternative approach relying on random compression of the feature vector combined with Gaussian process regression. The proposed approach is particularly motivated by the setting in which the response is conditionally independent of the features given the projection to a low dimensional manifold. Conditionally on the random compression matrix and a smoothness parameter, the posterior distribution for the regression surface and posterior predictive distributions are available analytically. Running the analysis in parallel for many random compression matrices and smoothness parameters, model averaging is used to combine the results. The algorithm can be implemented rapidly even in very large $p$ and moderately large $n$ nonparametric regression, has strong theoretical justification, and is found to yield state of the art predictive performance.

Graphical models express conditional independence relationships among variables. Although methods for vector-valued data are well established, functional data graphical models remain underdeveloped. By functional data, we refer to data that are realizations of random functions varying over a continuum (e.g., images, signals). We introduce a notion of conditional independence between random functions, and construct a framework for Bayesian inference of undirected, decomposable graphs in the multivariate functional data context. This framework is based on extending Markov distributions and hyper Markov laws from random variables to random processes, providing a principled alternative to naive application of multivariate methods to discretized functional data. Markov properties facilitate the composition of likelihoods and priors according to the decomposition of a graph. Our focus is on Gaussian process graphical models using orthogonal basis expansions. We propose a hyper-inverse-Wishart-process prior for the covariance kernels of the infinite coefficient sequences of the basis expansion, and establish its existence and uniqueness. We also prove the strong hyper Markov property and the conjugacy of this prior under a finite rank condition of the prior kernel parameter. Stochastic search Markov chain Monte Carlo algorithms are developed for posterior inference, assessed through simulations, and applied to a study of brain activity and alcoholism.

We propose an algorithm that removes the visually unpleasant effects of cradling in X-ray images of panel paintings, with the goal of improving the X-ray image readability by art experts. The algorithm consists of three stages. In the first stage the location of the cradle is detected automatically and the grayscale inconsistency, caused by the thickness of the cradle, is corrected. In a second stage we use a method called morphological component analysis to separate the X-ray image into a so-called cartoon part and a texture part, where the latter contains mostly the wood grain from both the panel and the cradling. The algorithm next learns a Bayesian factor model that distinguishes between the texture patterns that originate from the cradle and those from other components such as the panel and/or the painting on the panel surface, and finally uses this to remove the textures associated with the cradle. We apply the algorithm to a number of historically important paintings on panel. We also show how it can be used to digitally remove stretcher artifacts from X-rays of paintings on canvas. We compare our results with those obtained manually by best current practices in art conservation as well as on a ground truth dataset, consisting of X-ray images of a painting before and after removal of the physically attached cradle.

The promise of Bayesian methods for big data sets has not fully been realized due to the lack of scalable computational algorithms. For massive data, it is necessary to store and process subsets on different machines in a distributed manner. We propose a simple, general, and highly efficient approach, which first runs a posterior sampling algorithm in parallel on different machines for subsets of a large data set. To combine these subset posteriors, we calculate the Wasserstein barycenter via a highly efficient linear program. The resulting estimate for the Wasserstein posterior (WASP) has an atomic form, facilitating straightforward estimation of posterior summaries of functionals of interest. The WASP approach allows posterior sampling algorithms for smaller data sets to be trivially scaled to huge data. We provide theoretical justification in terms of posterior consistency and algorithm efficiency. Examples are provided in complex settings including Gaussian process regression and nonparametric Bayes mixture models.

Learning of low dimensional structure in multidimensional data is a canonical problem in machine learning. One common approach is to suppose that the observed data are close to a lower-dimensional smooth manifold. There are a rich variety of manifold learning methods available, which allow mapping of data points to the manifold. However, there is a clear lack of probabilistic methods that allow learning of the manifold along with the generative distribution of the observed data. The best attempt is the Gaussian process latent variable model (GP-LVM), but identifiability issues lead to poor performance. We solve these issues by proposing a novel Coulomb repulsive process (Corp) for locations of points on the manifold, inspired by physical models of electrostatic interactions among particles. Combining this process with a GP prior for the mapping function yields a novel electrostatic GP (electroGP) process. Focusing on the simple case of a one-dimensional manifold, we develop efficient inference algorithms, and illustrate substantially improved performance in a variety of experiments including filling in missing frames in video.

The modern scale of data has brought new challenges to Bayesian inference. In particular, conventional MCMC algorithms are computationally very expensive for large data sets. A promising approach to solve this problem is embarrassingly parallel MCMC (EP-MCMC), which first partitions the data into multiple subsets and runs independent sampling algorithms on each subset. The subset posterior draws are then aggregated via some combining rules to obtain the final approximation. Existing EP-MCMC algorithms are limited by approximation accuracy and difficulty in resampling. In this article, we propose a new EP-MCMC algorithm PART that solves these problems. The new algorithm applies random partition trees to combine the subset posterior draws, which is distribution-free, easy to resample from and can adapt to multiple scales. We provide theoretical justification and extensive experiments illustrating empirical performance.

Variable screening is a fast dimension reduction technique for assisting high dimensional feature selection. As a preselection method, it selects a moderate size subset of candidate variables for further refining via feature selection to produce the final model. The performance of variable screening depends on both computational efficiency and the ability to dramatically reduce the number of variables without discarding the important ones. When the data dimension $p$ is substantially larger than the sample size $n$, variable screening becomes crucial as 1) Faster feature selection algorithms are needed; 2) Conditions guaranteeing selection consistency might fail to hold.This article studies a class of linear screening methods and establishes consistency theory for this special class. In particular, we prove the restricted diagonally dominant (RDD) condition is a necessary and sufficient condition for strong screening consistency. As concrete examples, we show two screening methods $SIS$ and $HOLP$ are both strong screening consistent (subject to additional constraints) with large probability if $n > O((\rho s + \sigma/\tau)^2\log p)$ under random designs. In addition, we relate the RDD condition to the irrepresentable condition, and highlight limitations of $SIS$.

MOTIVATION: Both single marker and simultaneous analysis face challenges in GWAS due to the large number of markers genotyped for a small number of subjects. This large p small n problem is particularly challenging when the trait under investigation has low heritability. METHOD: In this article, we propose a two-stage approach that is a hybrid method of single and simultaneous analysis designed to improve genomic prediction of complex traits. In the first stage, we use a Bayesian independent screening method to select the most promising SNPs. In the second stage, we rely on a hierarchical model to analyze the joint impact of the selected markers. The model is designed to take into account familial dependence in the different subjects, while using local-global shrinkage priors on the marker effects. RESULTS: We evaluate the performance in simulation studies, and consider an application to animal breeding data. The illustrative data analysis reveals an encouraging result in terms of prediction performance and computational cost.

Time-varying adjacency matrices encoding the presence or absence of a relation among entities are available in many research fields. Motivated by an application to studying dynamic networks among sports teams, we propose a Bayesian nonparametric model. The proposed approach uses a logistic mapping from the probability matrix, encoding link probabilities between each team, to an embedded latent relational space. Within this latent space, we incorporate a dictionary of Gaussian process (GP) latent trajectories characterizing changes over time in each team, while allowing learning of the number of latent dimensions through a specially tailored prior for the GP covariance. The model is provably flexible and borrows strength across the network and over time. We provide simulation experiments and an application to the Italian soccer Championship.

We introduce an algorithm that removes the deleterious effect of cradling on X-ray images of paintings on wooden panels. The algorithm consists of a three stage procedure. Firstly, the cradled regions are located automatically. The second step consists of separating the X-ray image into a textural and image component. In the last step the algorithm learns to distinguish between the texture caused by the wooden cradle and the texture belonging to the original painted wooden panel. The results obtained with our method are compared with those obtained manually by best current practice.

Many Bayesian learning methods for massive data benefit from working with small subsets of observations. In particular, significant progress has been made in scalable Bayesian learning via stochastic approximation. However, Bayesian learning methods in distributed computing environments are often problem- or distribution-specific and use ad hoc techniques. We propose a novel general approach to Bayesian inference that is scalable and robust to corruption in the data. Our technique is based on the idea of splitting the data into several non-overlapping subgroups, evaluating the posterior distribution given each independent subgroup, and then combining the results. The main novelty is the proposed aggregation step which is based on finding the geometric median of posterior distributions. We present both theoretical and numerical results illustrating the advantages of our approach.

We present a scalable Bayesian framework for low-rank decomposition of multiway tensor data with missing observations. The key issue of pre-specifying the rank of the decomposition is sidestepped in a principled manner using a multiplicative gamma process prior. Both continuous and binary data can be analyzed under the framework, in a coherent way using fully conjugate Bayesian analysis. In particular, the analysis in the non-conjugate binary case is facilitated via the use of the Pólya-Gamma sampling strategy which elicits closed-form Gibbs sampling updates. The resulting samplers are efficient and enable us to apply our framework to large-scale problems, with time-complexity that is linear in the number of observed entries in the tensor. This is especially attractive in analyzing very large but sparsely observed tensors with very few known entries. Moreover, our method admits easy extension to the supervised setting where entities in one or more tensor modes have labels. Our method outperforms several state-of-the-art tensor decomposition methods on various synthetic and benchmark real-world datasets.

MOTIVATION: Estimating a phenotype distribution conditional on a set of discrete-valued predictors is a commonly encountered task. For example, interest may be in how the density of a quantitative trait varies with single nucleotide polymorphisms and patient characteristics. The subset of important predictors is not usually known in advance. This becomes more challenging with a high-dimensional predictor set when there is the possibility of interaction. RESULTS: We demonstrate a novel non-parametric Bayes method based on a tensor factorization of predictor-dependent weights for Gaussian kernels. The method uses multistage predictor selection for dimension reduction, providing succinct models for the phenotype distribution. The resulting conditional density morphs flexibly with the selected predictors. In a simulation study and an application to molecular epidemiology data, we demonstrate advantages over commonly used methods.

There is increasing interest in broad application areas in defining flexible joint models for data having a variety of measurement scales, while also allowing data of complex types, such as functions, images and documents. We consider a general framework for nonparametric Bayes joint modeling through mixture models that incorporate dependence across data types through a joint mixing measure. The mixing measure is assigned a novel infinite tensor factorization (ITF) prior that allows flexible dependence in cluster allocation across data types. The ITF prior is formulated as a tensor product of stick-breaking processes. Focusing on a convenient special case corresponding to a Parafac factorization, we provide basic theory justifying the flexibility of the proposed prior. Focusing on ITF mixtures of product kernels, we develop a new Gibbs sampling algorithm for routine implementation relying on slice sampling. The methods are compared with alternative joint mixture models based on Dirichlet processes and related approaches through simulations and real data applications.

Bayesian classification commonly relies on probit models, with data augmentation algorithms used for posterior computation. By imputing latent Gaussian variables, one can often trivially adapt computational approaches used in Gaussian models. However, MCMC for multinomial probit (MNP) models can be inefficient in practice due to high posterior dependence between latent variables and parameters, and to difficulties in efficiently sampling latent variables when there are more than two categories. To address these problems, we propose a new class of diagonal orthant (DO) multinomial models. The key characteristics of these models include conditional independence of the latent variables given model parameters, avoidance of arbitrary identifiability restrictions, and simple expressions for category probabilities. We show substantially improved computational efficiency and comparable predictive performance to MNP.

In modeling multivariate time series, it is important to allow time-varying smoothness in the mean and covariance process. In particular, there may be certain time intervals exhibiting rapid changes and others in which changes are slow. If such locally adaptive smoothness is not accounted for, one can obtain misleading inferences and predictions, with over-smoothing across erratic time intervals and under-smoothing across times exhibiting slow variation. This can lead to miscalibration of predictive intervals, which can be substantially too narrow or wide depending on the time. We propose a continuous multivariate stochastic process for time series having locally varying smoothness in both the mean and covariance matrix. This process is constructed utilizing latent dictionary functions in time, which are given nested Gaussian process priors and linearly related to the observed data through a sparse mapping. Using a differential equation representation, we bypass usual computational bottlenecks in obtaining MCMC and online algorithms for approximate Bayesian inference. The performance is assessed in simulations and illustrated in a financial application.

Nonparametric estimation of the conditional distribution of a response given high-dimensional features is a challenging problem. It is important to allow not only the mean but also the variance and shape of the response density to change flexibly with features, which are massive-dimensional. We propose a multiscale dictionary learning model, which expresses the conditional response density as a convex combination of dictionary densities, with the densities used and their weights dependent on the path through a tree decomposition of the feature space. A fast graph partitioning algorithm is applied to obtain the tree decomposition, with Bayesian methods then used to adaptively prune and average over different sub-trees in a soft probabilistic manner. The algorithm scales efficiently to approximately one million features. State of the art predictive performance is demonstrated for toy examples and two neuroscience applications including up to a million features.

MOTIVATION: In biomedical research a growing number of platforms and technologies are used to measure diverse but related information, and the task of clustering a set of objects based on multiple sources of data arises in several applications. Most current approaches to multisource clustering either independently determine a separate clustering for each data source or determine a single 'joint' clustering for all data sources. There is a need for more flexible approaches that simultaneously model the dependence and the heterogeneity of the data sources. RESULTS: We propose an integrative statistical model that permits a separate clustering of the objects for each data source. These separate clusterings adhere loosely to an overall consensus clustering, and hence they are not independent. We describe a computationally scalable Bayesian framework for simultaneous estimation of both the consensus clustering and the source-specific clusterings. We demonstrate that this flexible approach is more robust than joint clustering of all data sources, and is more powerful than clustering each data source independently. We present an application to subtype identification of breast cancer tumor samples using publicly available data from The Cancer Genome Atlas. AVAILABILITY: R code with instructions and examples is available at http://people.duke.edu/%7Eel113/software.html.

Unsupervised multilayered (“deep”) models are considered for imagery. The model is represented using a hierarchical convolutional factor-analysis construction, with sparse factor loadings and scores. The computation of layer-dependent model parameters is implemented within a Bayesian setting, employing a Gibbs sampler and variational Bayesian (VB) analysis that explicitly exploit the convolutional nature of the expansion. To address large-scale and streaming data, an online version of VB is also developed. The number of dictionary elements at each layer is inferred from the data, based on a beta-Bernoulli implementation of the Indian buffet process. Example results are presented for several image-processing applications, with comparisons to related models in the literature.

We propose a multiresolution Gaussian process to capture long-range, non-Markovian dependencies while allowing for abrupt changes. The multiresolution GP hierarchically couples a collection of smooth GPs, each defined over an element of a random nested partition. Long-range dependencies are captured by the top-level GP while the partition points define the abrupt changes. Due to the inherent conjugacy of the GPs, one can analytically marginalize the GPs and compute the conditional likelihood of the observations given the partition tree. This allows for efficient inference of the partition itself, for which we employ graph-theoretic techniques. We apply the multiresolution GP to the analysis of Magnetoencephalography (MEG) recordings of brain activity.

Discrete mixtures are used routinely in broad sweeping applications ranging from unsupervised settings to fully supervised multi-task learning. Indeed, finite mixtures and infinite mixtures, relying on Dirichlet processes and modifications, have become a standard tool. One important issue that arises in using discrete mixtures is low separation in the components; in particular, different components can be introduced that are very similar and hence redundant. Such redundancy leads to too many clusters that are too similar, degrading performance in unsupervised learning and leading to computational problems and an unnecessarily complex model in supervised settings. Redundancy can arise in the absence of a penalty on components placed close together even when a Bayesian approach is used to learn the number of components. To solve this problem, we propose a novel prior that generates components from a repulsive process, automatically penalizing redundant components. We characterize this repulsive prior theoretically and propose a Markov chain Monte Carlo sampling algorithm for posterior computation. The methods are illustrated using synthetic examples and an iris data set.

Nonparametric Bayesian methods are considered for recovery of imagery based upon compressive, incomplete, and/or noisy measurements. A truncated beta-Bernoulli process is employed to infer an appropriate dictionary for the data under test and also for image recovery. In the context of compressive sensing, significant improvements in image recovery are manifested using learned dictionaries, relative to using standard orthonormal image expansions. The compressive-measurement projections are also optimized for the learned dictionary. Additionally, we consider simpler (incomplete) measurements, defined by measuring a subset of image pixels, uniformly selected at random. Spatial interrelationships within imagery are exploited through use of the Dirichlet and probit stick-breaking processes. Several example results are presented, with comparisons to other methods in the literature.

A dependent hierarchical beta process (dHBP) is developed as a prior for data that may be represented in terms of a sparse set of latent features (dictionary elements), with covariate dependent feature usage. The dHBP is applicable to general covariates and data models, imposing that signals with similar covariates are likely to be manifested in terms of similar features. As an application, we consider the simultaneous sparse modeling of multiple images, with the covariate of a given image linked to its similarity to all other images (as applied in manifold learning). Efficient inference is performed using hybrid Gibbs, Metropolis-Hastings and slice sampling.

In recent years, a rich variety of shrinkage priors have been proposed that have great promise in addressing massive regression problems. In general, these new priors can be expressed as scale mixtures of normals, but have more complex forms and better properties than traditional Cauchy and double exponential priors. We first propose a new class of normal scale mixtures through a novel generalized beta distribution that encompasses many interesting priors as special cases. This encompassing framework should prove useful in comparing competing priors, considering properties and revealing close connections. We then develop a class of variational Bayes approximations through the new hierarchy presented that will scale more efficiently to the types of truly massive data sets that are now encountered routinely.

A new Le ́vy process prior is proposed for an uncountable collection of covariate- dependent feature-learning measures; the model is called the kernel beta process (KBP). Available covariates are handled efficiently via the kernel construction, with covariates assumed observed with each data sample (“customer”), and latent covariates learned for each feature (“dish”). Each customer selects dishes from an infinite buffet, in a manner analogous to the beta process, with the added constraint that a customer first decides probabilistically whether to “consider” a dish, based on the distance in covariate space between the customer and dish. If a customer does consider a particular dish, that dish is then selected probabilistically as in the beta process. The beta process is recovered as a limiting case of the KBP. An efficient Gibbs sampler is developed for computations, and state-of-the-art results are presented for image processing and music analysis tasks.

The nested Chinese restaurant process is extended to design a nonparametric topic-model tree for representation of human choices. Each tree branch corresponds to a type of person, and each node (topic) has a corresponding probability vector over items that may be selected. The observed data are assumed to have associated temporal covariates (corresponding to the time at which choices are made), and we wish to impose that with increasing time it is more probable that topics deeper in the tree are utilized. This structure is imposed by developing a new “change point" stick-breaking model that is coupled with a Poisson and product-of-gammas construction. To share topics across the tree nodes, topic distributions are drawn from a Dirichlet process. As a demonstration of this concept, we analyze real data on course selections of undergraduate students at Duke University, with the goal of uncovering and concisely representing structure in the curriculum and in the characteristics of the student body.

Protein-protein interactions (PPIs) are essential to most fundamental cellular processes. There has been increasing interest in reconstructing PPIs networks. However, several critical difficulties exist in obtaining reliable predictions. Noticeably, false positive rates can be as high as >80%. Error correction from each generating source can be both time-consuming and inefficient due to the difficulty of covering the errors from multiple levels of data processing procedures within a single test. We propose a novel Bayesian integration method, deemed nonparametric Bayes ensemble learning (NBEL), to lower the misclassification rate (both false positives and negatives) through automatically up-weighting data sources that are most informative, while down-weighting less informative and biased sources. Extensive studies indicate that NBEL is significantly more robust than the classic naïve Bayes to unreliable, error-prone and contaminated data. On a large human data set our NBEL approach predicts many more PPIs than naïve Bayes. This suggests that previous studies may have large numbers of not only false positives but also false negatives. The validation on two human PPIs datasets having high quality supports our observations. Our experiments demonstrate that it is feasible to predict high-throughput PPIs computationally with substantially reduced false positives and false negatives. The ability of predicting large numbers of PPIs both reliably and automatically may inspire people to use computational approaches to correct data errors in general, and may speed up PPIs prediction with high quality. Such a reliable prediction may provide a solid platform to other studies such as protein functions prediction and roles of PPIs in disease susceptibility.

We consider problems for which one has incomplete binary matrices that evolve with time (e.g., the votes of legislators on particular legislation, with each year characterized by a different such matrix). An objective of such analysis is to infer structure and inter-relationships underlying the matrices, here defined by latent features associated with each axis of the matrix. In addition, it is assumed that documents are available for the entities associated with at least one of the matrix axes. By jointly analyzing the matrices and documents, one may be used to inform the other within the analysis, and the model offers the opportunity to predict matrix values (e.g., votes) based only on an associated document (e.g., legislation). The research presented here merges two areas of machine-learning that have previously been investigated separately: incomplete-matrix analysis and topic modeling. The analysis is performed from a Bayesian perspective, with efficient inference constituted via Gibbs sampling. The framework is demonstrated by considering all voting data and available documents (legislation) during the 220-year lifetime of the United States Senate and House of Representatives.

BACKGROUND: Nonparametric Bayesian techniques have been developed recently to extend the sophistication of factor models, allowing one to infer the number of appropriate factors from the observed data. We consider such techniques for sparse factor analysis, with application to gene-expression data from three virus challenge studies. Particular attention is placed on employing the Beta Process (BP), the Indian Buffet Process (IBP), and related sparseness-promoting techniques to infer a proper number of factors. The posterior density function on the model parameters is computed using Gibbs sampling and variational Bayesian (VB) analysis. RESULTS: Time-evolving gene-expression data are considered for respiratory syncytial virus (RSV), Rhino virus, and influenza, using blood samples from healthy human subjects. These data were acquired in three challenge studies, each executed after receiving institutional review board (IRB) approval from Duke University. Comparisons are made between several alternative means of per-forming nonparametric factor analysis on these data, with comparisons as well to sparse-PCA and Penalized Matrix Decomposition (PMD), closely related non-Bayesian approaches. CONCLUSIONS: Applying the Beta Process to the factor scores, or to the singular values of a pseudo-SVD construction, the proposed algorithms infer the number of factors in gene-expression data. For real data the "true" number of factors is unknown; in our simulations we consider a range of noise variances, and the proposed Bayesian models inferred the number of factors accurately relative to other methods in the literature, such as sparse-PCA and PMD. We have also identified a "pan-viral" factor of importance for each of the three viruses considered in this study. We have identified a set of genes associated with this pan-viral factor, of interest for early detection of such viruses based upon the host response, as quantified via gene-expression data.

A Bayesian dynamic model is developed to model complex sequential data, with a focus on audio signals from music. The music is represented in terms of a sequence of discrete observations, and the sequence is modeled using a hidden Markov model (HMM) with time-evolving parameters. The model imposes the belief that observations that are temporally proximate are more likely to be drawn from HMMs with similar parameters, while also allowing for “innovation” associated with abrupt changes in the music texture. Segmentation of a given musical piece is constituted via the model inference and the results are compared with other models and also to a conventional music-theoretic analysis.

We propose a multi-task learning (MTL) framework for non-linear classification, based on an infinite set of local experts in feature space. The usage of local experts enables sharing at the expert-level, encouraging the borrowing of information even if tasks are similar only in subregions of feature space. A kernel stick-breaking process (KSBP) prior is imposed on the underlying distribution of class labels, so that the number of experts is inferred in the posterior and thus model selection issues are avoided. The MTL is implemented by imposing a Dirichlet process (DP) prior on a layer above the task-dependent KSBPs.

A non-parametric Bayesian model is proposed for processing multiple images. The analysis employs image features and, when present, the words associated with accompanying annotations. The model clusters the images into classes, and each image is segmented into a set of objects, also allowing the opportunity to assign a word to each object (localized labeling). Each object is assumed to be represented as a heterogeneous mix of components, with this realized via mixture models linking image features to object types. The number of image classes, number of object types, and the characteristics of the object-feature mixture models are inferred non-parametrically. To constitute spatially contiguous objects, a new logistic stick-breaking process is developed. Inference is performed efficiently via variational Bayesian analysis, with example results presented on two image databases.

The kernel stick-breaking process (KSBP) is employed to segment general imagery, imposing the condition that patches (small blocks of pixels) that are spatially proximate are more likely to be associated with the same cluster (segment). The number of clusters is not set a priori and is inferred from the hierarchical Bayesian model. Further, KSBP is integrated with a shared Dirichlet process prior to simultaneously model multiple images, inferring their inter-relationships. This latter application may be useful for sorting and learning relationships between multiple images. The Bayesian inference algorithm is based on a hybrid of variational Bayesian analysis and local sampling. In addition to providing details on the model and associated inference framework, example results are presented for several image-analysis problems.

Compressive sensing (CS) is an emerging £eld that, under appropriate conditions, can signi£cantly reduce the number of measurements required for a given signal. In many applications, one is interested in multiple signals that may be measured in multiple CS-type measurements, where here each signal corresponds to a sensing "task". In this paper we propose a novel multitask compressive sensing framework based on a Bayesian formalism, where a Dirichlet process (DP) prior is employed, yielding a principled means of simultaneously inferring the appropriate sharing mechanisms as well as CS inversion for each task. A variational Bayesian (VB) inference algorithm is employed to estimate the full posterior on the model parameters.

The dynamic hierarchical Dirichlet process (dHDP) is developed to model the time-evolving statistical properties of sequential data sets. The data collected at any time point are represented via a mixture associated with an appropriate underlying model, in the framework of HDP. The statistical properties of data collected at consecutive time points are linked via a random parameter that controls their probabilistic similarity. The sharing mechanisms of the time-evolving data are derived, and a relatively simple Markov Chain Monte Carlo sampler is developed. Experimental results are presented to demonstrate the model.

A new hierarchical nonparametric Bayesian model is proposed for the problem of multitask learning (MTL) with sequential data. Sequential data are typically modeled with a hidden Markov model (HMM), for which one often must choose an appropriate model structure (number of states) before learning. Here we model sequential data from each task with an infinite hidden Markov model (iHMM), avoiding the problem of model selection. The MTL for iHMMs is implemented by imposing a nested Dirichlet process (nDP) prior on the base distributions of the iHMMs. The nDP-iHMM MTL method allows us to perform task-level clustering and data-level clustering simultaneously, with which the learning for individual iHMMs is enhanced and between-task similarities are learned. Learning and inference for the nDP-iHMM MTL are based on a Gibbs sampler. The effectiveness of the framework is demonstrated using synthetic data as well as real music data.

In multi-task learning our goal is to design regression or classification models for each of the tasks and appropriately share information between tasks. A Dirichlet process (DP) prior can be used to encourage task clustering. However, the DP prior does not allow local clustering of tasks with respect to a subset of the feature vector without making independence assumptions. Motivated by this problem, we develop a new multitask-learning prior, termed the matrix stick-breaking process (MSBP), which encourages cross-task sharing of data. However, the MSBP allows separate clustering and borrowing of information for the different feature components. This is important when tasks are more closely related for certain features than for others. Bayesian inference proceeds by a Gibbs sampling algorithm and the approach is illustrated using a simulated example and a multi-national application.