An Information-theoretic Approach to Distribution Shifts

Patrick Forré, Ryota Tomioka, Marco Federici

Publication date: 7 June 2021

Abstract: Safely deploying machine learning models to the real world is often a challenging process. Models trained with data obtained from a specific geographic location tend to fail when queried with data obtained elsewhere, agents trained in a simulation can struggle to adapt when deployed in the real world or novel environments, and neural networks that are fit to a subset of the population might carry some selection bias into their decision process. In this work, we describe the problem of data shift from a novel information-theoretic perspective by (i) identifying and describing the different sources of error, (ii) comparing some of the most promising objectives explored in the recent domain generalization, and fair classification literature. From our theoretical analysis and empirical evaluation, we conclude that the model selection procedure needs to be guided by careful considerations regarding the observed data, the factors used for correction, and the structure of the data-generating process.

Has companion code repository: https://github.com/mfederici/dsit

This page was built for publication: An Information-theoretic Approach to Distribution Shifts

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6369676)