A Short Survey on Importance Weighting for Machine Learning

arXiv:2403.10175v1 Announce Type: new
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 …

abstract applications arxiv cs.ai cs.lg distribution example function importance instance machine machine learning probability sense simplicity statistics stat.ml survey type