ZOMA unifies hybrid zeroth-order estimators, bias corrections (GT/ED/EXTRA), and accelerations (STORM/PAGE/L2S) for decentralized nonconvex PL minimax optimization, claiming convergence rates matching centralized methods plus linear speedup.
Near-optimal decentralized momentum method for nonconvex-pl minimax problems
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
math.OC 2representative citing papers
DiMA's claimed O(kappa^2 epsilon^{-2}) communication complexity omits a (1-lambda)^{-3} factor that follows from the paper's own Eq. (14).
citing papers explorer
-
A Unified Zeroth-Order Approach for Decentralized Minimax Optimization
ZOMA unifies hybrid zeroth-order estimators, bias corrections (GT/ED/EXTRA), and accelerations (STORM/PAGE/L2S) for decentralized nonconvex PL minimax optimization, claiming convergence rates matching centralized methods plus linear speedup.
-
Communication-Efficient Decentralized Stochastic Minimax Optimization
DiMA's claimed O(kappa^2 epsilon^{-2}) communication complexity omits a (1-lambda)^{-3} factor that follows from the paper's own Eq. (14).