论文标题
与相关政策的多代理相互作用建模
Multi-Agent Interactions Modeling with Correlated Policies
论文作者
论文摘要
在多机构系统中，由于代理之间的高相关性，出现了复杂的相互作用行为。但是，先前关于从示威的建模多代理相互作用的工作主要是通过假设政策之间的独立性及其奖励结构来限制。在本文中，我们将多代理相互作用建模问题投入到多代理模仿学习框架中，并通过近似对手的策略进行明确的相关策略建模，这些策略可以恢复可以重新生成类似相互作用的代理政策。因此，我们开发了一种使用相关策略（CODAIL）的分散的对抗性模仿学习算法，该算法允许去中心化的培训和执行。各种实验表明，Codail可以更好地再重新生成靠近示威者的复杂相互作用，并且优于最先进的多代理模仿学习方法。我们的代码可在\ url {https://github.com/apexrl/codail}上获得。
In multi-agent systems, complex interacting behaviors arise due to the high correlations among agents. However, previous work on modeling multi-agent interactions from demonstrations is primarily constrained by assuming the independence among policies and their reward structures. In this paper, we cast the multi-agent interactions modeling problem into a multi-agent imitation learning framework with explicit modeling of correlated policies by approximating opponents' policies, which can recover agents' policies that can regenerate similar interactions. Consequently, we develop a Decentralized Adversarial Imitation Learning algorithm with Correlated policies (CoDAIL), which allows for decentralized training and execution. Various experiments demonstrate that CoDAIL can better regenerate complex interactions close to the demonstrators and outperforms state-of-the-art multi-agent imitation learning methods. Our code is available at \url{https://github.com/apexrl/CoDAIL}.
