arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.09882cs.LGcs.AI
收藏

软行动优先权:迈向强有力的政策转移

Soft Action Priors: Towards Robust Policy Transfer

Univ. Lille(里尔大学);CNRS(法国国家科学研究中心);Inria(法国国家信息与自动化研究所)

Matheus Centa, Philippe Preux

摘要:

尽管强化学习在许多具有挑战性的问题上取得了成功,但它仍然面临着样本效率低下的问题,这可以通过向智能体引入先验知识来缓解。然而,强化学习中的许多迁移技术都假设教师是专家。在本文中,我们使用来自强化学习的动作先验作为推理框架-也就是说,每个状态下的动作分布类似于教师策略,而不是贝叶斯先验-以恢复最先进的策略蒸馏技术。然后,我们提出了一类自适应方法,可以鲁棒地利用行动先验结合奖励塑造和辅助正则化损失。与以前的工作相比,我们开发的算法,利用次优的行动先验,但可能会传授有价值的知识-我们称之为软行动先验。所提出的算法通过调整教师反馈的强度来适应,根据教师在每个状态下的有用性的估计。我们进行了表格实验,结果表明,所提出的方法实现了最先进的性能,超越它时,从次优先验学习。最后,我们证明了自适应算法在连续动作深度RL问题中的鲁棒性,与现有的策略蒸馏方法相比,自适应算法大大提高了稳定性。

英文摘要:

Despite success in many challenging problems, reinforcement learning (RL) is still confronted with sample inefficiency, which can be mitigated by introducing prior knowledge to agents. However, many transfer techniques in reinforcement learning make the limiting assumption that the teacher is an expert. In this paper, we use the action prior from the Reinforcement Learning as Inference framework - that is, a distribution over actions at each state which resembles a teacher policy, rather than a Bayesian prior - to recover state-of-the-art policy distillation techniques. Then, we propose a class of adaptive methods that can robustly exploit action priors by combining reward shaping and auxiliary regularization losses. In contrast to prior work, we develop algorithms for leveraging suboptimal action priors that may nevertheless impart valuable knowledge - which we call soft action priors. The proposed algorithms adapt by adjusting the strength of teacher feedback according to an estimate of the teacher's usefulness in each state. We perform tabular experiments, which show that the proposed methods achieve state-of-the-art performance, surpassing it when learning from suboptimal priors. Finally, we demonstrate the robustness of the adaptive algorithms in continuous action deep RL problems, in which adaptive algorithms considerably improved stability when compared to existing policy distillation methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑