论文标题
Hilonet：分层模仿从非对准的观察中学习
HILONet: Hierarchical Imitation Learning from Non-Aligned Observations
论文作者
论文摘要
这是在非时期环境中仅仅从观察的轨迹中进行的挑战，因为大多数模仿学习方法旨在通过逐步进行演示来模仿专家。但是，在现实情况下很少能够对准演示。在这项工作中，我们提出了一种新的模仿学习方法，称为层次模仿从观察（Hilonet）学习，该方法采用了层次结构，从动态的观察结果中选择可行的子目标。我们的方法可以通过实现这些子目标来解决各种任务，无论它是否具有单个目标位置。我们还提出了提高层次结构样本效率的三种不同方法。我们使用多种环境进行了广泛的实验。结果表明表现和学习效率的提高。
It is challenging learning from demonstrated observation-only trajectories in a non-time-aligned environment because most imitation learning methods aim to imitate experts by following the demonstration step-by-step. However, aligned demonstrations are seldom obtainable in real-world scenarios. In this work, we propose a new imitation learning approach called Hierarchical Imitation Learning from Observation(HILONet), which adopts a hierarchical structure to choose feasible sub-goals from demonstrated observations dynamically. Our method can solve all kinds of tasks by achieving these sub-goals, whether it has a single goal position or not. We also present three different ways to increase sample efficiency in the hierarchical structure. We conduct extensive experiments using several environments. The results show the improvement in both performance and learning efficiency.
