Temporal abstraction refers to the ability of an agent to use behaviours of controllers which act for a limited, variable amount of time. The options framework describes such behaviours as consisting of a subset of states in which they can initiate, an internal policy and a stochastic termination condition. However, much of the subsequent work on option discovery has ignored the initiation set, because of difficulty in learning it from data. We provide a generalization of initiation sets suitable for general function approximation, by defining an interest function associated with an option. We derive a gradient-based learning algorithm for interest functions, leading to a new interest-option-critic architecture. We investigate how interest functions can be leveraged to learn interpretable and reusable temporal abstractions. We demonstrate the efficacy of the proposed approach through quantitative and qualitative results, in both discrete and continuous environments.
Temporal Abstraction in Reinforcement Learning with the Successor Representation
DeepMind（深度思维）；Alberta Machine Intelligence Institute (Amii)（阿尔伯塔机器智能研究所（Amii））；University of Alberta（阿尔伯塔大学）；Quebec AI Institute (Mila)（魁北克人工智能研究所（Mila））；McGill University（麦吉尔大学）
Multi-layer Abstraction for Nested Generation of Options (MANGO) in Hierarchical Reinforcement Learning
Human Inspired Technology Research Center, Università di Padova, Padova, PD 35121 IT（人类启发技术研究中心，帕多瓦大学，帕多瓦，PD 35121 IT）；Vrije Universiteit Amsterdam, Amsterdam, Netherlands（阿姆斯特丹自由大学，阿姆斯特丹，荷兰）
