We consider a machine learning setup where one training dataset is used to train multiple models across slightly different data distributions. This occurs when customized models are needed for various deployment environments. To reduce storage and training costs, we propose the DRSSS method, which combines distributionally robust (DR) optimization and safe sample screening (SSS). The key benefit of this method is that models trained on the reduced dataset will perform the same as those trained on the full dataset for all possible different environments. In this paper, we focus on covariate shift as a type of data distribution change and demonstrate the effectiveness of our method through experiments.
@article{arxiv.2406.05964,
title = {Distributionally Robust Safe Sample Elimination under Covariate Shift},
author = {Hiroyuki Hanada and Tatsuya Aoyama and Satoshi Akahane and Tomonari Tanaka and Yoshito Okura and Yu Inatsu and Noriaki Hashimoto and Shion Takeno and Taro Murayama and Hanju Lee and Shinya Kojima and Ichiro Takeuchi},
journal= {arXiv preprint arXiv:2406.05964},
year = {2024}
}