Utilising Gradient-Based Proposals within Sequential Monte Carlo Samplers for Training of Partial Bayesian Neural Networks
Year: 2026
Type: conference-paper
Abstract: Partial Bayesian neural networks (pBNNs) have been shown to perform competitively with fully Bayesian neural networks while only having a subset of the parameters be stochastic. Using sequential Monte Carlo (SMC) samplers as the inference method for pBNNs gives a non-parametric probabilistic estimation of the stochastic parameters, and has shown improved performance over parametric methods. We introduce a new SMC-based training method for pBNNs by utilising a guided proposal and incorporating gradient-based Markov kernels, which gives us better scalability on high dimensional problems. We show that our new method outperforms the state-of-the-art (SOTA) in terms of optimal loss.
Authors: Andrew Millard, Joshua Murphy, Zheng Zhao, Simon Maskell
Institutions: University of Liverpool, Linköping University
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
Funder: Royal Academy of Engineering
OpenAlex ID: W4416010240
FWCI:
Cites:
Cited by:
Related to: