English

JaxSGMC: Modular stochastic gradient MCMC in JAX

Computation 2025-05-19 v1 Applications Machine Learning

Abstract

We present JaxSGMC, an application-agnostic library for stochastic gradient Markov chain Monte Carlo (SG-MCMC) in JAX. SG-MCMC schemes are uncertainty quantification (UQ) methods that scale to large datasets and high-dimensional models, enabling trustworthy neural network predictions via Bayesian deep learning. JaxSGMC implements several state-of-the-art SG-MCMC samplers to promote UQ in deep learning by reducing the barriers of entry for switching from stochastic optimization to SG-MCMC sampling. Additionally, JaxSGMC allows users to build custom samplers from standard SG-MCMC building blocks. Due to this modular structure, we anticipate that JaxSGMC will accelerate research into novel SG-MCMC schemes and facilitate their application across a broad range of domains.

Cite

@article{arxiv.2505.11190,
  title  = {JaxSGMC: Modular stochastic gradient MCMC in JAX},
  author = {Stephan Thaler and Paul Fuchs and Ana Cukarska and Julija Zavadlav},
  journal= {arXiv preprint arXiv:2505.11190},
  year   = {2025}
}
R2 v1 2026-06-28T23:35:55.962Z