Data-Augmented Game Starts for Accelerating Self-Play Exploration in Imperfect Information Games

TL;DR

This paper introduces Data-Augmented Game Starts (DAGS), a method that uses offline human demonstration data to initialize reinforcement learning in complex imperfect-information games, significantly improving exploration and equilibrium approximation.

Contribution

The paper proposes DAGS, a novel initialization strategy for reinforcement learning in large-scale imperfect-information games, leveraging offline data to accelerate exploration and improve equilibrium quality.

Findings

DAGS achieves lower exploitability in synthetic and real games.

Augmenting starting states enhances exploration in challenging games.

Multi-task observation flags mitigate bias introduced by data augmentation.

Abstract

Finding approximate equilibria for large-scale imperfect-information competitive games such as StarCraft, Dota, and CounterStrike remains computationally infeasible due to sparse rewards and challenging exploration over long horizons. In this paper, we propose a multi-agent starting-state sampling strategy designed to substantially accelerate online exploration in regularized policy-gradient game methods for two-player zero-sum (2p0s) games. Motivated by an assumption that offline demonstrations from skilled humans can provide good coverage of high-level strategies relevant to equilibrium play, we propose the initialization of reinforcement learning data collection at intermediate states sampled from offline data to facilitate exploration of strategically relevant subgames. Referring to this method as Data-Augmented Game Starts (DAGS), we perform experiments using synthetic datasets and analytically tractable, long-horizon control variants of two-player Kuhn Poker, Goofspiel, and a counterexample game designed to penalize biased beliefs over hidden information. Under fixed computational budgets, DAGS enables regularized policy gradient methods to achieve lower exploitability in games with significantly more challenging exploration. We show that augmenting starting state distributions when solving imperfect information games can lead to biased equilibria, and we provide a straightforward mitigation to this in the form of multi-task observation flags. Finally, we release a new set of benchmark environments that drastically increase exploration challenges and state counts in existing OpenSpiel games while keeping exploitability measurements analytically tractable.