Learning Latent Representations to Influence Multi-Agent Interaction
Learning Latent Representations to Influence Multi-Agent Interaction
复制标题
DOI：
--
复制
发表时间：
2020-11
期刊：
影响因子：
--
通讯作者：
Annie Xie;Dylan P. Losey;R. Tolsma;Chelsea Finn;Dorsa Sadigh
中科院分区：
文献类型：
--
作者：
Annie Xie;Dylan P. Losey;R. Tolsma;Chelsea Finn;Dorsa Sadigh
与人类或机器人进行无障碍交互是很困难的，因为这些智能体是不稳定的。他们根据自我代理人的行为更新自己的策略，而自我代理人必须预测这些变化以适应。受人类的启发，我们认识到机器人不需要明确地对另一个代理将做出的每个低级动作进行建模;相反，我们可以通过高级表示来捕获其他代理的潜在策略。我们提出了一个基于强化学习的框架，用于学习代理策略的潜在表示，其中自我代理确定其行为与其他代理未来策略之间的关系。然后，自我代理人利用这些潜在的动态来影响其他代理人，有目的地引导他们采取适合共同适应的政策。在几个模拟域和现实世界的空气曲棍球比赛，我们的方法优于替代品，并学会影响其他代理。
Seamlessly interacting with humans or robots is hard because these agents are non-stationary. They update their policy in response to the ego agent's behavior, and the ego agent must anticipate these changes to co-adapt. Inspired by humans, we recognize that robots do not need to explicitly model every low-level action another agent will make; instead, we can capture the latent strategy of other agents through high-level representations. We propose a reinforcement learning-based framework for learning latent representations of an agent's policy, where the ego agent identifies the relationship between its behavior and the other agent's future strategy. The ego agent then leverages these latent dynamics to influence the other agent, purposely guiding them towards policies suitable for co-adaptation. Across several simulated domains and a real-world air hockey game, our approach outperforms the alternatives and learns to influence the other agent.