Benign overfitting
Estimators that fit noisy training data exactly and still predict well.
Stein's paradox and the James-Stein estimatorStein, 1956; James and Stein, 1961; Efron and Morris, 19738 questions
Double descentBelkin et al., 2019; Belkin et al., 20208 questions
Benign Overfitting in Linear RegressionBartlett et al., 2020; Tsigler and Bartlett, 20238 questions
Harmless interpolation of noisy data in regressionMuthukumar et al., 20208 questions
Surprises in High-Dimensional Ridgeless Least Squares InterpolationHastie et al., 20228 questions
Dimension free ridge regressionCheng and Montanari, 20248 questions
Uniform convergence of interpolatorsKoehler et al., 2021; Zhou et al., 20218 questions
Kernel interpolationLiang and Rakhlin, 2020; Belkin et al., 20198 questions
Benign, Tempered, or CatastrophicMallinar et al., 20228 questions
The implicit bias of gradient descentSoudry et al., 20188 questions