Current browse context:
stat.ML
Change to browse by:
References & Citations
Statistics > Machine Learning
Title: Challenges in Markov chain Monte Carlo for Bayesian neural networks
(Submitted on 15 Oct 2019 (v1), last revised 1 Oct 2021 (this version, v6))
Abstract: Markov chain Monte Carlo (MCMC) methods have not been broadly adopted in Bayesian neural networks (BNNs). This paper initially reviews the main challenges in sampling from the parameter posterior of a neural network via MCMC. Such challenges culminate to lack of convergence to the parameter posterior. Nevertheless, this paper shows that a non-converged Markov chain, generated via MCMC sampling from the parameter space of a neural network, can yield via Bayesian marginalization a valuable posterior predictive distribution of the output of the neural network. Classification examples based on multilayer perceptrons showcase highly accurate posterior predictive distributions. The postulate of limited scope for MCMC developments in BNNs is partially valid; an asymptotically exact parameter posterior seems less plausible, yet an accurate posterior predictive distribution is a tenable research avenue.
Submission history
From: Theodore Papamarkou [view email][v1] Tue, 15 Oct 2019 05:43:45 GMT (3253kb,D)
[v2] Tue, 22 Oct 2019 18:02:21 GMT (3267kb,D)
[v3] Thu, 14 Nov 2019 03:37:55 GMT (3267kb,D)
[v4] Tue, 23 Feb 2021 10:39:02 GMT (2792kb,D)
[v5] Wed, 11 Aug 2021 12:55:46 GMT (2791kb,D)
[v6] Fri, 1 Oct 2021 23:22:03 GMT (2791kb,D)
Link back to: arXiv, form interface, contact.