Diffusion Models for Offline Multi-agent Reinforcement Learning with Safety Constraints
This work addresses safety risks in MARL for real-world applications, representing an incremental improvement by integrating diffusion models into an existing CTDE architecture.
The paper tackles the problem of applying multi-agent reinforcement learning (MARL) to safety-critical scenarios by proposing a diffusion model-based framework for offline learning, which enhances safety and achieves superior performance on the DSRL benchmark compared to existing methods.
In recent advancements in Multi-agent Reinforcement Learning (MARL), its application has extended to various safety-critical scenarios. However, most methods focus on online learning, which presents substantial risks when deployed in real-world settings. Addressing this challenge, we introduce an innovative framework integrating diffusion models within the MARL paradigm. This approach notably enhances the safety of actions taken by multiple agents through risk mitigation while modeling coordinated action. Our framework is grounded in the Centralized Training with Decentralized Execution (CTDE) architecture, augmented by a Diffusion Model for prediction trajectory generation. Additionally, we incorporate a specialized algorithm to further ensure operational safety. We evaluate our model against baselines on the DSRL benchmark. Experiment results demonstrate that our model not only adheres to stringent safety constraints but also achieves superior performance compared to existing methodologies. This underscores the potential of our approach in advancing the safety and efficacy of MARL in real-world applications.