Electronic Arts SEED

Improving Sample Efficiency in Multi-Agent Reinforcement Learning for Simulated Football Games via Exploration

Advancing multi-agent reinforcement learning within simulated football environments requires balancing computational efficiency with the ability to master complex, strategic gameplay.

• Replacing LSTM units with four-layer multilayer perceptrons and fixed positional encodings reduced training time by 21.8%.
• Integrating a Random Network Distillation (RND) exploration bonus increased sample efficiency by 13.3% per training phase.
• RND proved more effective than self-supervised intrinsic reward mechanisms, which tended to overfit on rudimentary strategies like ball possession in complex environments.