Furthermore, from the proof of [41, Theorem 1], for all k ≥ 0 we get ∥Gxi(xk, yk)∥ ≤ (β + 2L)∥xk+1 i − xk i ∥, i = 1,

> 0, P k ≜ L(xk, yk) + 2 ρ2µ + 1 2ρ − 4( 1 ρ − L2 2µ ) ∥yk − yk−1∥2, which is the value of the potential function at iteration k ≥ 0 –here, c1, c2 > 0 follows from the conditio

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

browse 1 citing papers

representative citing papers

A Stochastic GDA Method With Backtracking For Solving Nonconvex Concave Minimax Problems

math.OC · 2024-03-12 · unverdicted · novelty 7.0

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

Showing 1 of 1 citing paper.

A Stochastic GDA Method With Backtracking For Solving Nonconvex Concave Minimax Problems math.OC · 2024-03-12 · unverdicted · none · ref 77
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.

Furthermore, from the proof of [41, Theorem 1], for all k ≥ 0 we get ∥Gxi(xk, yk)∥ ≤ (β + 2L)∥xk+1 i − xk i ∥, i = 1,

fields

years

verdicts

representative citing papers

citing papers explorer