基于人为驱动的动态数据集扩充改进行为克隆
Improving Behavioural Cloning with Human-Driven Dynamic Dataset Augmentation
摘要：
行为克隆已被广泛用于训练代理，并被认为是一种快速和坚实的方法来教授基于专家轨迹的一般行为。这种方法遵循监督学习范式，并且强烈依赖于数据的分布。在我们的论文中，我们展示了如何将行为克隆与人在环训练相结合，解决了它的一些缺陷，并提供了一个代理任务特定的纠正，以克服棘手的情况，同时加快训练时间，降低所需的资源。为此，我们引入了一种新的方法，允许专家在模拟过程中的任何时刻控制代理，并为其有问题的情况提供最佳解决方案。我们的实验表明，这种方法导致更好的政策，无论是在定量评估和人性化。
英文摘要：
Behavioural cloning has been extensively used to train agents and is recognized as a fast and solid approach to teach general behaviours based on expert trajectories. Such method follows the supervised learning paradigm and it strongly depends on the distribution of the data. In our paper, we show how combining behavioural cloning with human-in-the-loop training solves some of its flaws and provides an agent task-specific corrections to overcome tricky situations while speeding up the training time and lowering the required resources. To do this, we introduce a novel approach that allows an expert to take control of the agent at any moment during a simulation and provide optimal solutions to its problematic situations. Our experiments show that this approach leads to better policies both in terms of quantitative evaluation and in human-likeliness.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
