论文标题
在深度强化学习中学习可转移的概念
Learning Transferable Concepts in Deep Reinforcement Learning
论文作者
论文摘要
尽管人类和动物在一生中逐步学习并利用他们的经验来解决新任务，但标准的深入强化学习方法专门用于一次仅解决一项任务。结果，在新情况下，他们获得的信息几乎无法重复使用。在这里，我们介绍了有关利用先验知识解决未来任务的问题的新观点。我们表明，学习感官输入的离散表示可以提供高级抽象，该抽象在多个任务中很常见，从而促进信息的转移。特别是，我们表明，按照信息理论方法，可以通过自学来学习这种表示。我们的方法能够在机车和最佳控制任务中学习概念，从而提高已知和未知任务的样本效率，从而为具有概括能力赋予人造药物的新途径。
While humans and animals learn incrementally during their lifetimes and exploit their experience to solve new tasks, standard deep reinforcement learning methods specialize to solve only one task at a time. As a result, the information they acquire is hardly reusable in new situations. Here, we introduce a new perspective on the problem of leveraging prior knowledge to solve future tasks. We show that learning discrete representations of sensory inputs can provide a high-level abstraction that is common across multiple tasks, thus facilitating the transference of information. In particular, we show that it is possible to learn such representations by self-supervision, following an information theoretic approach. Our method is able to learn concepts in locomotive and optimal control tasks that increase the sample efficiency in both known and unknown tasks, opening a new path to endow artificial agents with generalization abilities.
