arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.05043cs.LGcs.RO
收藏

具有内在动机的深度强化学习目标条件策略学习

Learn Goal-Conditioned Policy with Intrinsic Motivation for Deep Reinforcement Learning

Zhejiang University(浙江大学);Westlake University(西湖大学);Westlake Institute for Advanced Study(西湖高等研究院)

Jinxin Liu, Donglin Wang, Qiangxing Tian, Zhengyu Chen

摘要:

学习一个广泛适用的、通用的、能够实现包括图像和文本描述在内的多种目标的策略对于智能体具有重要意义。考虑到这些感知特定的目标,深度强化学习研究的前沿是学习一个目标条件的政策,而无需手工制作的奖励。为了学习这种策略,最近的工作通常采取的奖励非参数距离的一个明确的嵌入空间中的给定目标。从不同的角度,我们提出了一种新的无监督学习方法命名为目标有条件的政策与内在动机(GPIM),共同学习的抽象级别的政策和目标有条件的政策。抽象级策略以潜在变量为条件,以优化策略,并发现进一步呈现为目标条件策略的感知特定目标的各种状态。学习的策略作为目标条件策略的内在奖励函数,以模仿抽象级策略诱导的轨迹。各种机器人任务的实验证明了我们提出的GPIM方法的有效性和效率,大大优于现有技术。

英文摘要:

It is of significance for an agent to learn a widely applicable and general-purpose policy that can achieve diverse goals including images and text descriptions. Considering such perceptually-specific goals, the frontier of deep reinforcement learning research is to learn a goal-conditioned policy without hand-crafted rewards. To learn this kind of policy, recent works usually take as the reward the non-parametric distance to a given goal in an explicit embedding space. From a different viewpoint, we propose a novel unsupervised learning approach named goal-conditioned policy with intrinsic motivation (GPIM), which jointly learns both an abstract-level policy and a goal-conditioned policy. The abstract-level policy is conditioned on a latent variable to optimize a discriminator and discovers diverse states that are further rendered into perceptually-specific goals for the goal-conditioned policy. The learned discriminator serves as an intrinsic reward function for the goal-conditioned policy to imitate the trajectory induced by the abstract-level policy. Experiments on various robotic tasks demonstrate the effectiveness and efficiency of our proposed GPIM method which substantially outperforms prior techniques.

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。

↑