arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2109.05490cs.LGcs.AI
收藏

HyAR:通过混合动作表示解决离散-连续动作强化学习

HyAR: Addressing Discrete-Continuous Action Reinforcement Learning via Hybrid Action Representation

College of Intelligence and Computing, Tianjin University(天津大学智能与计算学部);School of Cybersecurity, Northwestern Polytechnical University(西北工业大学网络空间安全学院)

Boyan Li, Hongyao Tang, Yan Zheng, Jianye Hao, Pengyi Li, Zhen Wang, Zhaopeng Meng, Li Wang

摘要:

离散-连续混合动作空间是机器人控制、游戏人工智能等许多实际问题的自然背景。然而,大多数强化学习(RL)的工作只展示了成功的控制与离散或连续的动作空间,而很少考虑到混合的动作空间。解决混合动作RL的一种简单方法是通过离散化或连续化将混合动作空间转换为统一的齐次动作空间,以便可以应用常规RL算法。然而,这忽略了混合动作空间的底层结构,也导致了可扩展性问题和额外的近似困难,从而导致退化的结果。在本文中,我们提出了混合动作表示(HyAR)学习一个紧凑的和可解码的潜在的表示空间的原始混合动作空间。HyAR构造了隐空间,并通过嵌入表和条件变分自动编码器(VAE)嵌入离散动作和连续参数之间的依赖关系。为了进一步提高有效性,通过无监督的环境动态预测来训练动作表示以使其在语义上平滑。最后,代理然后学习其政策与传统的DRL算法在学习的表示空间和与环境进行交互,通过解码的混合动作嵌入到原始的动作空间。我们评估HyAR在各种环境中的离散连续的动作空间。结果表明,HyAR的优越性相比,以前的基线,特别是对于高维动作空间。

英文摘要:

Discrete-continuous hybrid action space is a natural setting in many practical problems, such as robot control and game AI. However, most previous Reinforcement Learning (RL) works only demonstrate the success in controlling with either discrete or continuous action space, while seldom take into account the hybrid action space. One naive way to address hybrid action RL is to convert the hybrid action space into a unified homogeneous action space by discretization or continualization, so that conventional RL algorithms can be applied. However, this ignores the underlying structure of hybrid action space and also induces the scalability issue and additional approximation difficulties, thus leading to degenerated results. In this paper, we propose Hybrid Action Representation (HyAR) to learn a compact and decodable latent representation space for the original hybrid action space. HyAR constructs the latent space and embeds the dependence between discrete action and continuous parameter via an embedding table and conditional Variantional Auto-Encoder (VAE). To further improve the effectiveness, the action representation is trained to be semantically smooth through unsupervised environmental dynamics prediction. Finally, the agent then learns its policy with conventional DRL algorithms in the learned representation space and interacts with the environment by decoding the hybrid action embeddings to the original action space. We evaluate HyAR in a variety of environments with discrete-continuous action space. The results demonstrate the superiority of HyAR when compared with previous baselines, especially for high-dimensional action spaces.

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。

↑