Averaging the iterates of the stochastic order oracle algorithm yields asymptotically normal errors with covariance d/((d-1)^2 α^2) ∇²f(x*)^{-2}, tighter than the non-averaged version.
Dueling convex optimization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Ruppert-Polyak averaging for Stochastic Order Oracle
Averaging the iterates of the stochastic order oracle algorithm yields asymptotically normal errors with covariance d/((d-1)^2 α^2) ∇²f(x*)^{-2}, tighter than the non-averaged version.