Improved upper bound of Õ(ε^{-4/(3p+1)}) p-th order oracle complexity for convex-concave minimax problems via Monteiro-Svaiter acceleration, with matching lower bound Ω(ε^{-2/(3p-1)}).
Generalized optimistic methods for convex-concave saddle point problems
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
math.OC 2verdicts
UNVERDICTED 2representative citing papers
SGDA-B is the first backtracking-enabled stochastic GDA algorithm for nonconvex-concave minimax problems that achieves the best known complexity bounds among methods agnostic to L, μ, and σ².
citing papers explorer
-
Solving Convex-Concave Problems with $\tilde{\mathcal{O}}(\epsilon^{-4/(3p+1)})$ $p$th-Order Oracle Complexity
Improved upper bound of Õ(ε^{-4/(3p+1)}) p-th order oracle complexity for convex-concave minimax problems via Monteiro-Svaiter acceleration, with matching lower bound Ω(ε^{-2/(3p-1)}).
-
A Stochastic GDA Method With Backtracking For Solving Nonconvex Concave Minimax Problems
SGDA-B is the first backtracking-enabled stochastic GDA algorithm for nonconvex-concave minimax problems that achieves the best known complexity bounds among methods agnostic to L, μ, and σ².