Single-pass SGD with exponentially decaying steps is claimed to reach minimax-optimal excess risk in high-dimensional kernel regression for well-specified problems, with averaging handling misspecified problems.
Benign overfitting in linear regression
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
stat.ML 1years
2025 1verdicts
REJECT 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Learning Curves of Stochastic Gradient Descent in Kernel Regression
Single-pass SGD with exponentially decaying steps is claimed to reach minimax-optimal excess risk in high-dimensional kernel regression for well-specified problems, with averaging handling misspecified problems.