The paper considers matrix games with d actions per player.
Observed opponent actions enable optimal current-strategy convergence in matrix games
Zhang derives an efficient algorithm whose played strategies approach equilibrium at the minimax-optimal rate, up to logarithmic factors.
Academic
Yuheng Zhang
University of Illinois Urbana-Champaign
Research Digest··2 min read
Yuheng Zhang studies unknown two-player zero-sum games where each round reveals the sampled actions and one noisy payoff.
Why this paper
From University of Illinois Urbana-Champaign
In one line
An algorithm achieves minimax optimal last-iterate duality gap O~(√(d/t)) for zero-sum matrix games with observed opponent actions.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§