Fast Rates for Nonstationary Weighted Risk Minimization

ArXi:2602.05742v2 Announce Type: replace-cross Weighted empirical risk minimization is a common approach to prediction under distribution drift. This article studies its out-of-sample prediction error under nonstationarity. We provide a general decomposition of the excess risk into a learning term and an error term associated with distribution drift, and prove oracle inequalities for the learning error under mixing conditions.