STAS: Spatial-Temporal Return Decomposition for Multi-agent Reinforcement Learning
This addresses a specific bottleneck in cooperative multi-agent systems for scenarios with delayed rewards, representing an incremental improvement over existing methods.
The paper tackles the challenge of credit assignment in episodic multi-agent reinforcement learning, where global rewards are delayed, by introducing STAS, a method that decomposes global returns temporally and redistributes credit spatially using Shapley Value approximations, achieving superior performance over state-of-the-art baselines in MPE environments.
Centralized Training with Decentralized Execution (CTDE) has been proven to be an effective paradigm in cooperative multi-agent reinforcement learning (MARL). One of the major challenges is credit assignment, which aims to credit agents by their contributions. While prior studies have shown great success, their methods typically fail to work in episodic reinforcement learning scenarios where global rewards are revealed only at the end of the episode. They lack the functionality to model complicated relations of the delayed global reward in the temporal dimension and suffer from inefficiencies. To tackle this, we introduce Spatial-Temporal Attention with Shapley (STAS), a novel method that learns credit assignment in both temporal and spatial dimensions. It first decomposes the global return back to each time step, then utilizes the Shapley Value to redistribute the individual payoff from the decomposed global reward. To mitigate the computational complexity of the Shapley Value, we introduce an approximation of marginal contribution and utilize Monte Carlo sampling to estimate it. We evaluate our method on an Alice & Bob example and MPE environments across different scenarios. Our results demonstrate that our method effectively assigns spatial-temporal credit, outperforming all state-of-the-art baselines.