S Sagawa, A Raghunathan, PW Koh… - arXiv e-prints, 2020 - ui.adsabs.harvard.edu
We study why overparameterization--increasing model size well beyond the point of zero training error--can hurt test error on minority groups despite improving average test error …
S Sagawa, A Raghunathan… - … on Machine Learning, 2020 - proceedings.mlr.press
We study why overparameterization—increasing model size well beyond the point of zero training error—can hurt test error on minority groups despite improving average test error …
S Sagawa, A Raghunathan, PW Koh… - arXiv preprint arXiv …, 2020 - arxiv.org
We study why overparameterization--increasing model size well beyond the point of zero training error--can hurt test error on minority groups despite improving average test error …
S Sagawa, A Raghunathan, PW Koh, P Liang, A Singh… - pages.cs.wisc.edu
An Investigation of Why Overparameterization Exacerbates Spurious Correlation Page 1 An Investigation of Why Overparameterization Exacerbates Spurious Correlation Authors …
S Sagawa, A Raghunathan, PW Koh… - Proceedings of the 37th …, 2020 - dl.acm.org
We study why overparameterization--increasing model size well beyond the point of zero training error--can hurt test error on minority groups despite improving average test error …
S Sagawa, A Raghunathan, PW Koh, P Liang, A Singh… - pages.cs.wisc.edu
An Investigation of Why Overparameterization Exacerbates Spurious Correlation Page 1 An Investigation of Why Overparameterization Exacerbates Spurious Correlation Authors …