Recent progress in Quality Diversity Reinforcement Learning (QD-RL) has enabled learning a collection of behaviorally diverse, high performing policies. However, these methods typically involve storing thousands of policies, which results in high space-complexity and poor scaling to additional behaviors. Condensing the archive into a single model while retaining the performance and coverage of the original collection of policies has proved challenging. In this work, we propose using diffusion models to distill the archive into a single generative model over policy parameters. We show that our method achieves a compression ratio of 13x while recovering 98% of the original rewards and 89% of the original coverage. Further, the conditioning mechanism of diffusion models allows for flexibly selecting and sequencing behaviors, including using language. Project website: https://sites.google.com/view/policydiffusion/home

翻译：近期质量多样性强化学习（QD-RL）的进展使得学习一组行为多样化且高性能的策略成为可能。然而，此类方法通常需要存储数千个策略，导致高空间复杂度及随行为增加而扩展性变差。将档案压缩至单一模型同时保持原始策略集合的性能与覆盖范围被证明具有挑战性。本研究提出利用扩散模型将档案蒸馏为单一策略参数生成模型。实验表明，该方法在恢复98%原始奖励与89%原始覆盖范围的同时，实现了13倍的压缩比。此外，扩散模型的条件机制支持灵活选择与排序行为，甚至可通过语言进行控制。项目网站：https://sites.google.com/view/policydiffusion/home