Concept extraction using temporal-difference network EUROCON2009

摘要

In this paper, we propose a novel framework to extract temporally extended concepts in a grid world environment using a probable data structure named temporal-difference network. First a reinforcement-learning agent tries to learn its environment for the task of wall following. After that we train a newly introduced temporal-difference network (TDN) in the brain of the agent in order to gain a predictive model of the environment. At last the most promising sequences of action-observation of the given environment will be sorted out based on their probability.