Home

Improved Marginal Unbiased Score Expansion (MUSE) via Implicit...

Author
Marius Millea
Category
stat.ML
Date Published
2022/09/21
Date Retrieved
2022/09/22
Date Updated
2022/09/22
Description
We apply the technique of implicit differentiation to boost performance, reduce numerical error, and remove required user-tuning in the Marginal Unbiased Score Expansion (MUSE) algorithm for hierarchical Bayesian inference. We demonstrate these improvements on three representative inference problems: 1) an extended Neal's funnel 2) Bayesian neural networks, and 3) probabilistic principal component analysis. On our particular test cases, MUSE with implicit differentiation is faster than Hamiltonian Monte Carlo by factors of 155, 397, and 5, respectively, or factors of 65, 278, and 1 without implicit differentiation, and yields good approximate marginal posteriors. The Julia and Python MUSE packages have been updated to use implicit differentiation, and can solve problems defined by hand or with any of a number of popular probabilistic programming languages and automatic differentiation backends.
Posts
5
Readers
0
Score
1.25
Tweeters
5
URL
https://arxiv.org/abs/2209.10512
TOP