Non-Volatile Memory Accelerated Posterior Estimation
This addresses the challenge of overconfidence in machine learning models for researchers by making previously infeasible posterior approximations feasible.
The paper tackles the problem of approximating large posterior distributions in Bayesian inference by leveraging high-capacity persistent storage, enabling improved predictions in downstream tasks.
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.