Deterministic power-law stepsizes accelerate last-iterate convergence of Extragradient for biaffine min-max optimization to O(T^{-2/3+ε}) or O(T^{-1+ε}).
Zihan Zhang, Jason Lee, Simon Du, and Yuxin Chen
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
math.OC 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Optimal (N-1)-step fixed-point algorithms correspond exactly to (N-1)! arc diagrams that support composition, decomposition, and H-duality while producing new quasi-anytime optimal methods.
citing papers explorer
-
Accelerating Min-Max Optimization via Power-Law Stepsizes
Deterministic power-law stepsizes accelerate last-iterate convergence of Extragradient for biaffine min-max optimization to O(T^{-2/3+ε}) or O(T^{-1+ε}).
-
A Theory of Composition and Duality of Extremal Optimal Fixed-Point Algorithms
Optimal (N-1)-step fixed-point algorithms correspond exactly to (N-1)! arc diagrams that support composition, decomposition, and H-duality while producing new quasi-anytime optimal methods.