无需在线交互，通过双层优化恢复专家行为的奖励函数。
BiCQL-ML: A Bi-Level Conservative Q-Learning Framework for Maximum Likelihood Inverse Reinforcement Learning
- 双层框架联合优化奖励函数与保守Q值，避免显式策略学习。
- 在标准基准上提升奖励恢复精度与下游策略性能。
- 适合追求高可靠性奖励学习的离线强化学习研究者。
离线逆强化学习（IRL）旨在仅使用固定演示数据恢复解释专家行为的奖励函数，无需额外在线交互。我们提出BiCQL-ML，一种无需策略的离线IRL算法，采用双层框架联合优化奖励函数与保守Q函数，从而避免显式策略学习。该方法交替执行：(i) 在当前奖励下通过保守Q学习（CQL）学习保守Q函数；(ii) 更新奖励参数以最大化专家动作的期望Q值，同时抑制对分布外动作的过拟合。该过程可视为软值匹配原则下的最大似然估计。我们提供了理论保证，证明BiCQL-ML收敛至使专家策略为软最优的奖励函数。实验表明，在标准离线强化学习基准上，相比现有离线IRL基线，BiCQL-ML在奖励恢复和下游策略性能方面均有提升。
原文摘要 · Abstract (English)
Offline inverse reinforcement learning (IRL) aims to recover a reward function that explains expert behavior using only fixed demonstration data, without any additional online interaction. We propose BiCQL-ML, a policy-free offline IRL algorithm that jointly optimizes a reward function and a conservative Q-function in a bi-level framework, thereby avoiding explicit policy learning. The method alternates between (i) learning a conservative Q-function via Conservative Q-Learning (CQL) under the current reward, and (ii) updating the reward parameters to maximize the expected Q-values of expert actions while suppressing over-generalization to out-of-distribution actions. This procedure can be viewed as maximum likelihood estimation under a soft value matching principle. We provide theoretical guarantees that BiCQL-ML converges to a reward function under which the expert policy is soft-optimal. Empirically, we show on standard offline RL benchmarks that BiCQL-ML improves both reward recovery and downstream policy performance compared to existing offline IRL baselines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。