VR-SZD, a variance-reduced zeroth-order method using orthogonal structured directions, achieves O(d n^{2/3} ε^{-1}) function evaluations for non-convex composite finite-sum problems and linear convergence under the Polyak-Łojasiewicz condition, matching state-of-the-art rates at lower…
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2025 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
A Structured Proximal Stochastic Variance Reduced Zeroth-order Algorithm
VR-SZD, a variance-reduced zeroth-order method using orthogonal structured directions, achieves O(d n^{2/3} ε^{-1}) function evaluations for non-convex composite finite-sum problems and linear convergence under the Polyak-Łojasiewicz condition, matching state-of-the-art rates at lower…