Fixed-bandwidth Gaussian kernels let any spectral algorithm achieve minimax nonparametric rates, and the resulting two-stage procedure attains near-optimal transfer learning under concept shift.
Transfer learning for nonparametric classification: Minimax rate and adaptive classifier
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
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Fixed-Gaussian Spectral Algorithms: Minimax Optimal Rates for Misspecified Learning and Transfer
Fixed-bandwidth Gaussian kernels let any spectral algorithm achieve minimax nonparametric rates, and the resulting two-stage procedure attains near-optimal transfer learning under concept shift.