For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
A Study of Qualitative Knowledge-Based Exploration for Continuous Deep Reinforcement Learning
Chenxi LI Lei CAO Xiaoming LIU Xiliang CHEN Zhixiong XU Yongliang ZHANG
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2017/11/01
Online ISSN: 1745-1361
Type of Manuscript: LETTER
Category: Artificial Intelligence, Data Mining
deep reinforcement learning, knowledge, exploration strategy, cloud control systems,
Full Text: PDF(478KB)
>>Buy this Article
As an important method to solve sequential decision-making problems, reinforcement learning learns the policy of tasks through the interaction with environment. But it has difficulties scaling to large-scale problems. One of the reasons is the exploration and exploitation dilemma which may lead to inefficient learning. We present an approach that addresses this shortcoming by introducing qualitative knowledge into reinforcement learning using cloud control systems to represent ‘if-then’ rules. We use it as the heuristics exploration strategy to guide the action selection in deep reinforcement learning. Empirical evaluation results show that our approach can make significant improvement in the learning process.