arXiv research

This work proposes a method to learn Sparse Structured Ensembles using SG-MCMC and weight pruning.

New samplers improve MCMC efficiency in high dimensions.

Improved Bayesian inference for neuronal ensemble inference reduces computational cost.

A new stopping rule based on E-values helps efficiently use sampling in Bayesian Deep Ensembles.

This paper reviews multiple try MCMC methods for signal processing.

ESS improves MCMC efficiency for correlated & multimodal distributions.

Bayesian inference for factorial hidden Markov models is challenging due to the exponentially sized latent variable space. Standard Monte Carlo samplers can have difficulties effectively exploring the posterior landscape and are often restricted to exploration around localised regions that depend on initialisation. We …

Rodent hippocampal population codes represent important spatial information about the environment during navigation. Several computational methods have been developed to uncover the neural representation of spatial topology embedded in rodent hippocampal ensemble spike activity. Here we extend our previous work and pro…

Data assimilation for subsurface flow using latent diffusion models shows that ensemble Kalman methods may overestimate posterior uncertainty, while Monte Carlo sampling is more reliable.

Machine learning speeds up LSM parameter optimization and uncertainty quantification.

Paper assesses adversarial robustness of MCMC and BDK methods for deep Bayesian networks.

Bayesian neural networks struggle with accuracy and uncertainty quantification in complex models.

We study the computational complexity of Markov chain Monte Carlo (MCMC) methods for high-dimensional Bayesian linear regression under sparsity constraints. We first show that a Bayesian approach can achieve variable-selection consistency under relatively mild conditions on the design matrix. We then demonstrate that t…

Paper introduces XBART for nonlinear regression, outperforming XGBoost.

Bayesian learning made scalable with posteriors library.

SBMC method improves uncertainty estimation in deep learning models.

We propose a novel sampling framework for inference in probabilistic models: an active learning approach that converges more quickly (in wall-clock time) than Markov chain Monte Carlo (MCMC) benchmarks. The central challenge in probabilistic inference is numerical integration, to average over ensembles of models or unk…

Bayesian spatial predictive synthesis improves spatial data predictions.

SVI and GP surrogates improve calibration of ABMs in epidemiology.

Bayesian CART models improve insurance claims frequency prediction and interpretation.

Unified framework for MCMC algorithms simplifies their design and application.

Subsampling MCMC speeds up Bayesian inference for large datasets.

JaxSGMC simplifies SG-MCMC for Bayesian deep learning.

Unified framework for MCMC and machine learning problems.

Develops a new framework to understand MCMC dynamics as flows on Wasserstein space.

Paper proposes first unlearning algorithm for MCMC models.

Improves MCMC and VI methods by optimising simulation bias.

The paper explores a non-convergent MCMC method for EBM learning.

TensorFlow Probability MCMC toolkit improves MCMC efficiency for modern hardware.

Stochastic EM with biased MCMC improves inference stability.

Monte Carlo (MC) methods are widely used for Bayesian inference and optimization in statistics, signal processing and machine learning. A well-known class of MC methods are Markov Chain Monte Carlo (MCMC) algorithms. In order to foster better exploration of the state space, specially in high-dimensional applications, s…

Cyclical MCMC tackles high-dimensional multimodal distributions, showing convergence under certain conditions.

Develops Cyclical SG-MCMC for exploring multimodal posterior distributions in deep learning.

Paper compares MCMC-based copula methods for exchange option pricing.

Improves MCMC sampling for multi-modal distributions.

Nonlinear MCMC improves Bayesian machine learning sampling.

Underdamped Langevin MCMC achieves faster convergence than overdamped methods.

Bayesian optimization uses BNNs as efficient surrogate models for expensive function evaluations.

Particle MCMC involves using a particle filter within an MCMC algorithm. For inference of a model which involves an unobserved stochastic process, the standard implementation uses the particle filter to propose new values for the stochastic process, and MCMC moves to propose new values for the parameters. We show how p…

New analysis of SGD with MCMC gradient estimator shows convergence rate and saddle point escape.

Stochastic gradient MCMC (SG-MCMC) algorithms have proven useful in scaling Bayesian inference to large datasets under an assumption of i.i.d data. We instead develop an SG-MCMC algorithm to learn the parameters of hidden Markov models (HMMs) for time-dependent data. There are two challenges to applying SG-MCMC in this…

Proposes MIVI for efficient posterior estimation and design of MCMC transitions.

A new method optimizes particles in SG-MCMC for better approximation.

Infinite BART model selects number of trees and allows different functions for clusters.

This study evaluates different normalizing flow architectures for MCMC.

AutoStep MCMC adapts step size locally for better sampling efficiency.

Markov chain Monte Carlo (MCMC) is a popular and successful general-purpose tool for Bayesian inference. However, MCMC cannot be practically applied to large data sets because of the prohibitive cost of evaluating every likelihood term at every iteration. Here we present Firefly Monte Carlo (FlyMC) an auxiliary variabl…

Stochastic gradient MCMC (SG-MCMC) has played an important role in large-scale Bayesian learning, with well-developed theoretical convergence properties. In such applications of SG-MCMC, it is becoming increasingly popular to employ distributed systems, where stochastic gradients are computed based on some outdated par…