SPGM is a first-order method for smooth convex optimization that is dynamically optimal at each step via subgame perfect equilibrium, outperforming OGM in preliminary tests.
Universal gradient methods for convex optimization problems.Mathematical Programming, 152(1):381–404
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Beyond Minimax Optimality: A Subgame Perfect Gradient Method
SPGM is a first-order method for smooth convex optimization that is dynamically optimal at each step via subgame perfect equilibrium, outperforming OGM in preliminary tests.