Post-hoc loss-calibration for Bayesian neural networks

06/13/2021
by   Meet P. Vadera, et al.
0

Bayesian decision theory provides an elegant framework for acting optimally under uncertainty when tractable posterior distributions are available. Modern Bayesian models, however, typically involve intractable posteriors that are approximated with, potentially crude, surrogates. This difficulty has engendered loss-calibrated techniques that aim to learn posterior approximations that favor high-utility decisions. In this paper, focusing on Bayesian neural networks, we develop methods for correcting approximate posterior predictive distributions encouraging them to prefer high-utility decisions. In contrast to previous work, our approach is agnostic to the choice of the approximate inference algorithm, allows for efficient test time decision making through amortization, and empirically produces higher quality decisions. We demonstrate the effectiveness of our approach through controlled experiments spanning a diversity of tasks and datasets.

READ FULL TEXT

page 5

page 9

research
01/10/2022

Loss-calibrated expectation propagation for approximate Bayesian decision-making

Approximate Bayesian inference methods provide a powerful suite of tools...
research
09/11/2019

Correcting Predictions for Approximate Bayesian Inference

Bayesian models quantify uncertainty and facilitate optimal decision-mak...
research
06/15/2022

On Calibrated Model Uncertainty in Deep Learning

Estimated uncertainty by approximate posteriors in Bayesian neural netwo...
research
07/17/2022

Uncertainty Calibration in Bayesian Neural Networks via Distance-Aware Priors

As we move away from the data, the predictive uncertainty should increas...
research
10/28/2020

Expressive yet Tractable Bayesian Deep Learning via Subnetwork Inference

The Bayesian paradigm has the potential to solve some of the core issues...
research
05/10/2018

Loss-Calibrated Approximate Inference in Bayesian Neural Networks

Current approaches in approximate inference for Bayesian neural networks...
research
04/29/2022

Tractable Uncertainty for Structure Learning

Bayesian structure learning allows one to capture uncertainty over the c...

Please sign up or login with your details

Forgot password? Click here to reset