219 citations · 307 across the 3 of their papers we have counts for
Showing cs.LGShow all
2 papers · 1 filter
cs.LG2020★ 5 cited
Selective Classification Can Magnify Disparities Across Groups
Erik Jones, Shiori Sagawa, Pang Wei Koh +2
Selective classification, in which models can abstain on uncertain predictions, is a natural approach to improving accuracy in settings where errors are costly but abstentions are…
cs.LG2020★ 83 cited
An Investigation of Why Overparameterization Exacerbates Spurious Correlations
Shiori Sagawa, Aditi Raghunathan, Pang Wei Koh +1
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 err…