Is MC Dropout Bayesian?
MC Dropout is a mainstream "free lunch" method in medical imaging for approximate Bayesian computations (ABC). Its appeal is to solve out-of-the-box the daunting task of ABC and uncertainty quantification in Neural Networks (NNs); to fall within the variational inference (VI) framework; an...
Saved in:
Main Authors: | , , , , , , , , , |
---|---|
Format: | Journal Article |
Language: | English |
Published: |
08-10-2021
|
Subjects: | |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | MC Dropout is a mainstream "free lunch" method in medical imaging for
approximate Bayesian computations (ABC). Its appeal is to solve out-of-the-box
the daunting task of ABC and uncertainty quantification in Neural Networks
(NNs); to fall within the variational inference (VI) framework; and to propose
a highly multimodal, faithful predictive posterior. We question the properties
of MC Dropout for approximate inference, as in fact MC Dropout changes the
Bayesian model; its predictive posterior assigns $0$ probability to the true
model on closed-form benchmarks; the multimodality of its predictive posterior
is not a property of the true predictive posterior but a design artefact. To
address the need for VI on arbitrary models, we share a generic VI engine
within the pytorch framework. The code includes a carefully designed
implementation of structured (diagonal plus low-rank) multivariate normal
variational families, and mixtures thereof. It is intended as a go-to
no-free-lunch approach, addressing shortcomings of mean-field VI with an
adjustable trade-off between expressivity and computational complexity. |
---|---|
DOI: | 10.48550/arxiv.2110.04286 |