With advances in reinforcement learning (RL), agents are now being developed in high-stakes application domains such as healthcare and transportation. Explaining the behavior of these agents is challenging, as the environments in which they act have large state spaces, and their decision-making can be affected by delayed rewards, making it difficult to analyze their behavior. To address this problem, several approaches have been developed. Some approaches attempt to convey the global behavior of the agent, describing the actions it takes in different states. Other approaches devised local explanations which provide information regarding the agent's decision-making in a particular state. In this paper, we combine global and local explanation methods, and evaluate their joint and separate contributions, providing (to the best of our knowledge) the first user study of combined local and global explanations for RL agents. Specifically, we augment strategy summaries that extract important trajectories of states from simulations of the agent with saliency maps which show what information the agent attends to. Our results show that the choice of what states to include in the summary (global information) strongly affects people's understanding of agents: participants shown summaries that included important states significantly outperformed participants who were presented with agent behavior in a set of world-states that are likely to appear during gameplay. We find mixed results with respect to augmenting demonstrations with saliency maps (local information), as the addition of saliency maps, in the form of raw heat maps, did not significantly improve performance in most cases. However, we do find some evidence that saliency maps can help users better understand what information the agent relies on during its decision-making, suggesting avenues for future work that can further improve explanations of RL agents.

随着强化学习 (RL) 的进步，智能体现在正在高风险的应用领域中开发，例如医疗保健和运输。解释这些智能体的行为具有挑战性，因为它们所处的环境具有很大的状态空间，而且它们的决策可能会受到延迟奖励的影响，因此很难分析它们的行为。为了解决这个问题，已经开发了几种方法。一些方法试图传达代理的全局行为，描述它在不同状态下采取的行动。本地设计的其他方法提供有关代理在特定状态下的决策信息的解释。在本文中，我们结合了全局和局部解释方法，并评估了它们联合和单独的贡献，（据我们所知）提供了第一个对 RL 代理的局部和全局解释相结合的用户研究。具体来说，我们增加了策略摘要，从代理的模拟中提取重要的状态轨迹，显着图显示代理关注的信息。我们的结果表明，选择将哪些状态包含在摘要（全局信息）中会强烈影响人们对代理的理解：参与者展示的包含重要状态的摘要的表现明显优于在游戏过程中可能出现的一组世界状态中呈现代理行为的参与者。我们发现在使用显着图（本地信息）增强演示方面的结果好坏参半，因为在大多数情况下，以原始热图的形式添加显着图并没有显着提高性能。然而，我们确实发现了一些证据，表明显着图可以帮助用户更好地了解代理在其决策过程中所依赖的信息，为未来的工作提供了途径，可以进一步改进对 RL 代理的解释。以原始热图的形式，在大多数情况下并没有显着提高性能。然而，我们确实发现了一些证据，表明显着图可以帮助用户更好地了解代理在其决策过程中所依赖的信息，为未来的工作提供了途径，可以进一步改进对 RL 代理的解释。以原始热图的形式，在大多数情况下并没有显着提高性能。然而，我们确实发现了一些证据，表明显着图可以帮助用户更好地了解代理在其决策过程中所依赖的信息，为未来的工作提供了途径，可以进一步改进对 RL 代理的解释。