On Feynman--Kac training of partial Bayesian neural networks

arXiv:2310.19608v2 Announce Type: replace
Abstract: Recently, partial Bayesian neural networks (pBNNs), which only consider a subset of the parameters to be stochastic, were shown to perform competitively with full Bayesian neural networks. However, pBNNs are often multi-modal in the latent variable space and thus challenging to approximate with parametric models. To address this problem, we propose an efficient sampling-based training strategy, wherein the training of a pBNN is formulated as simulating a Feynman--Kac model. We then describe variations of sequential …

abstract arxiv bayesian cs.lg feynman modal multi-modal networks neural networks parameters parametric space stat.ml stochastic training type