Bayesian Neural Networks with Domain Knowledge Priors

arXiv:2402.13410v1 Announce Type: new
Abstract: Bayesian neural networks (BNNs) have recently gained popularity due to their ability to quantify model uncertainty. However, specifying a prior for BNNs that captures relevant domain knowledge is often extremely challenging. In this work, we propose a framework for integrating general forms of domain knowledge (i.e., any knowledge that can be represented by a loss function) into a BNN prior through variational inference, while enabling computationally efficient posterior inference and sampling. Specifically, our approach results …

abstract arxiv bayesian cs.lg domain domain knowledge forms framework general knowledge networks neural networks prior stat.ml type uncertainty work