BNNpriors: A library for Bayesian neural network inference with different prior distributions
Abstract
Bayesian neural networks have shown great promise in many applications where calibrated uncertainty estimates are crucial and can often also lead to a higher predictive performance. However, it remains challenging to choose a good prior distribution over their weights. While isotropic Gaussian priors are often chosen in practice due to their simplicity, they do not reflect our true prior beliefs well and can lead to suboptimal performance. Our new library, BNNpriors, enables state-of-the-art Markov Chain Monte Carlo inference on Bayesian neural networks with a wide range of predefined priors, including heavy-tailed ones, hierarchical ones, and mixture priors. Moreover, it follows a modular approach that eases the design and implementation of new custom priors. It has facilitated foundational discoveries on the nature of the cold posterior effect in Bayesian neural networks and will hopefully catalyze future research as well as practical applications in this area.
Cite
@article{arxiv.2105.06964,
title = {BNNpriors: A library for Bayesian neural network inference with different prior distributions},
author = {Vincent Fortuin and Adrià Garriga-Alonso and Mark van der Wilk and Laurence Aitchison},
journal= {arXiv preprint arXiv:2105.06964},
year = {2021}
}
Comments
Accepted for publication at Software Impacts