Learning-based approaches, such as reinforcement and imitation learning are gaining popularity in decision-making for autonomous driving. However, learned policies often fail to generalize and cannot handle novel situations well. Asking and answering questions in the form of "Would a policy perform well if the other agents had behaved differently?" can shed light on whether a policy has seen similar situations during training and generalizes well. In this work, a counterfactual policy evaluation is introduced that makes use of counterfactual worlds - worlds in which the behaviors of others are non-actual. If a policy can handle all counterfactual worlds well, it either has seen similar situations during training or it generalizes well and is deemed to be fit enough to be executed in the actual world. Additionally, by performing the counterfactual policy evaluation, causal relations and the influence of changing vehicle's behaviors on the surrounding vehicles becomes evident. To validate the proposed method, we learn a policy using reinforcement learning for a lane merging scenario. In the application-phase, the policy is only executed after the counterfactual policy evaluation has been performed and if the policy is found to be safe enough. We show that the proposed approach significantly decreases the collision-rate whilst maintaining a high success-rate.

中文翻译：

---

自动驾驶决策的反事实政策评估

基于学习的方法，例如强化学习和模仿学习，在自动驾驶决策中越来越受欢迎。然而，学习到的策略往往不能泛化，不能很好地处理新情况。以“如果其他代理的行为不同，政策会表现良好吗？”的形式提出和回答问题。可以阐明策略在训练过程中是否遇到过类似情况并很好地概括。在这项工作中，引入了一种反事实政策评估，它利用了反事实世界——其他人的行为是非实际的世界。如果一个策略可以很好地处理所有反事实世界，那么它要么在训练过程中遇到过类似的情况，要么很好地概括并被认为足够适合在现实世界中执行。此外，通过执行反事实策略评估，因果关系以及改变车辆行为对周围车辆的影响变得明显。为了验证所提出的方法，我们使用强化学习来学习用于车道合并场景的策略。在应用阶段，只有在执行了反事实策略评估并且发现策略足够安全后，才会执行策略。我们表明，所提出的方法显着降低了碰撞率，同时保持了高成功率。仅在执行反事实策略评估并且发现该策略足够安全后，才会执行该策略。我们表明，所提出的方法显着降低了碰撞率，同时保持了高成功率。仅在执行反事实策略评估并且发现该策略足够安全后，才会执行该策略。我们表明，所提出的方法显着降低了碰撞率，同时保持了高成功率。