Multigroup Robustness

arXiv:2405.00614v1 Announce Type: new
Abstract: To address the shortcomings of real-world datasets, robust learning algorithms have been designed to overcome arbitrary and indiscriminate data corruption. However, practical processes of gathering data may lead to patterns of data corruption that are localized to specific partitions of the training dataset. Motivated by critical applications where the learned model is deployed to make predictions about people from a rich collection of overlapping subpopulations, we initiate the study of multigroup robust algorithms whose robustness …

abstract algorithms applications arxiv corruption cs.lg data dataset datasets however patterns practical processes robust robustness training type world