arXiv research

Bayesian networks with latent variables are characterized and their likelihoods compared.

A new two-step MH method for Bayesian EL computation.

Proposes OBS, a method to adaptively combine Bayesian models online.

Improves Bayesian predictive performance in misspecified models.

R package for Bayesian empirical likelihood sampling using HMC.

Bayesian predictive inference analyzes a dataset to make predictions about new observations. When a model does not match the data, predictive accuracy suffers. We develop population empirical Bayes (POP-EB), a hierarchical framework that explicitly models the empirical population distribution as part of Bayesian analys…

VIB balances empirical and Bayesian approaches in predictive models.

Bayesian method infers contextual bandit policies robustly.

Herding and kernel herding are deterministic methods of choosing samples which summarise a probability distribution. A related task is choosing samples for estimating integrals using Bayesian quadrature. We show that the criterion minimised when selecting samples in kernel herding is equivalent to the posterior varianc…

Herding and kernel herding are deterministic methods of choosing samples which summarise a probability distribution. A related task is choosing samples for estimating integrals using Bayesian quadrature. We show that the criterion minimised when selecting samples in kernel herding is equivalent to the posterior varianc…

Deep ensembles mimic Bayesian averaging with learned priors.

Paper derives PAC-Bayesian bounds for LTI systems learning from empirical data.

A new clustering method using Bayesian techniques improves robustness and interpretability.

Adaptive Bayesian quadrature (ABQ) is a powerful approach to numerical integration that empirically compares favorably with Monte Carlo integration on problems of medium dimensionality (where non-adaptive quadrature is not competitive). Its key ingredient is an acquisition function that changes as a function of previou…

FP-BMA improves generalization by encouraging flat posteriors in Bayesian Model Averaging.

Stochastic variational inference for Bayesian deep neural network (DNN) requires specifying priors and approximate posterior distributions over neural network weights. Specifying meaningful weight priors is a challenging problem, particularly for scaling variational inference to deeper architectures involving high dime…

Study examines dependence properties of Bayesian neural network units in finite-width networks.

metabeta uses neural networks to speed up Bayesian mixed-effects regression.

Study improves Bayesian optimisation with ensemble transfer learning.

Bayesian optimization improves molecule design by addressing three pitfalls.

Transformer pretraining yields strong EB performance without explicit adaptation.

Empirical analysis serves as an important complement to theoretical analysis for studying practical Bayesian optimization. Often empirical insights expose strengths and weaknesses inaccessible to theoretical analysis. We define two metrics for comparing the performance of Bayesian optimization methods and propose a ran…

We study the problem of learning Bayesian network structures from data. We develop an algorithm for finding the k-best Bayesian network structures. We propose to compute the posterior probabilities of hypotheses of interest by Bayesian model averaging over the k-best Bayesian networks. We present empirical results on s…

This paper improves learning uncertain Bayesian networks from incomplete data.

Bayesian approach fixes overconfidence in ReLU networks, even slightly.

We prove semi-empirical concentration inequalities for random variables which are given as possibly nonlinear functions of independent random variables. These inequalities describe concentration of random variable in terms of the data/distribution-dependent Efron-Stein (ES) estimate of its variance and they do not requ…

Empirical study compares finite- and infinite-width BNNs, revealing performance differences under model mismatch.

Parallelized Bayesian quadrature improves sample efficiency and inference.

Bayesian neural networks can be partially stochastic without losing predictive power.

Bayesian optimization usually assumes that a Bayesian prior is given. However, the strong theoretical guarantees in Bayesian optimization are often regrettably compromised in practice because of unknown parameters in the prior. In this paper, we adopt a variant of empirical Bayes and show that, by estimating the Gaussi…

New algorithm optimizes PAC-Bayes bound without surrogate loss.

Proposes a new generalization bound for Bayesian deep nets without strict assumptions.

Bayesian Empirical Bayes extends EB to complex structures using probabilistic symmetry.

IDS improves sparse linear bandits by balancing information and regret.

Bayesian framework uses AI-generated data to improve parameter estimation.

New bound relaxes uniform gradient norm assumptions for PAC-Bayesian bounds.

Bayesian method learns from aggregated quantile data.

The optimal approach is to theorize after examining data, not before.

GBEST model improves survival analysis for small datasets.

DLM for BNNs fails to improve over ELBO optimization.

BFNs use Bayesian inference and neural networks for generative modeling.

Bayesian network structure learning is often performed in a Bayesian setting, by evaluating candidate structures using their posterior probabilities for a given data set. Score-based algorithms then use those posterior probabilities as an objective function and return the maximum a posteriori network as the learned mod…

Bayesian coresets improve scalable Bayesian inference.

This work introduces a new data-driven estimator for the Bayesian Cramér-Rao bound using score matching.

This paper connects masked pre-training to Bayesian model selection.

Novel hybrid method for Bayesian network structure learning reduces computational time without sacrificing accuracy.

We show that the multi-class support vector machine (MSVM) proposed by Lee et. al. (2004), can be viewed as a MAP estimation procedure under an appropriate probabilistic interpretation of the classifier. We also show that this interpretation can be extended to a hierarchical Bayesian architecture and to a fully-Bayesia…

Bayesian principles improve neural additive models for better feature selection and uncertainty.