Agent-Temporal Credit Assignment for Optimal Policy Preservation in Sparse Multi-Agent Reinforcement Learning
Kapoor, Aditya, Swamy, Sushant, Tessera, Kale-ab, Baranwal, Mayank, Sun, Mingfei, Khadilkar, Harshad, Albrecht, Stefano V.
–arXiv.org Artificial Intelligence
In multi-agent environments, agents often struggle to learn optimal policies due to sparse or delayed global rewards, particularly in long-horizon tasks where it is challenging to evaluate actions at intermediate time steps. We introduce Temporal-Agent Reward Redistribution (TAR$^2$), a novel approach designed to address the agent-temporal credit assignment problem by redistributing sparse rewards both temporally and across agents. TAR$^2$ decomposes sparse global rewards into time-step-specific rewards and calculates agent-specific contributions to these rewards. We theoretically prove that TAR$^2$ is equivalent to potential-based reward shaping, ensuring that the optimal policy remains unchanged. Empirical results demonstrate that TAR$^2$ stabilizes and accelerates the learning process. Additionally, we show that when TAR$^2$ is integrated with single-agent reinforcement learning algorithms, it performs as well as or better than traditional multi-agent reinforcement learning methods.
arXiv.org Artificial Intelligence
Dec-19-2024
- Genre:
- Overview (0.88)
- Research Report > New Finding (0.34)
- Industry:
- Leisure & Entertainment > Games > Computer Games (0.46)