A Short Survey on Importance Weighting for Machine Learning

TL;DR

This survey reviews the broad applications of importance weighting in machine learning, emphasizing its role in addressing distribution shift and improving statistical properties.

Contribution

It provides a comprehensive summary of importance weighting techniques and their applications across various machine learning tasks.

Findings

Importance weighting helps correct distribution shift in supervised learning.

It guarantees desirable statistical properties under certain assumptions.

The survey highlights diverse applications and future research directions.

Abstract

Importance weighting is a fundamental procedure in statistics and machine learning that weights the objective function or probability distribution based on the importance of the instance in some sense. The simplicity and usefulness of the idea has led to many applications of importance weighting. For example, it is known that supervised learning under an assumption about the difference between the training and test distributions, called distribution shift, can guarantee statistically desirable properties through importance weighting by their density ratio. This survey summarizes the broad applications of importance weighting in machine learning and related research.