Back to Search Start Over

Non-Volatile Memory Accelerated Posterior Estimation

Authors :
Wood, Andrew
Hershcovitch, Moshik
Waddington, Daniel
Cohen, Sarel
Chin, Peter
Publication Year :
2022

Abstract

Bayesian inference allows machine learning models to express uncertainty. Current machine learning models use only a single learnable parameter combination when making predictions, and as a result are highly overconfident when their predictions are wrong. To use more learnable parameter combinations efficiently, these samples must be drawn from the posterior distribution. Unfortunately computing the posterior directly is infeasible, so often researchers approximate it with a well known distribution such as a Gaussian. In this paper, we show that through the use of high-capacity persistent storage, models whose posterior distribution was too big to approximate are now feasible, leading to improved predictions in downstream tasks.

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.2202.10522
Document Type :
Working Paper