The Divergence of Reinforcement Learning Algorithms with Value-Iteration and Function Approximation

Abstract

This paper gives specific divergence examples of value-iteration for several major Reinforcement Learning and Adaptive Dynamic Programming algorithms, when using a function approximator for the value function. These divergence examples differ from previous divergence examples in the literature, in that they are applicable for a greedy policy, i.e. in a "value iteration" scenario. Perhaps surprisingly, with a greedy policy, it is also possible to get divergence for the algorithms TD(1) and Sarsa(1). In addition to these divergences, we also achieve divergence for the Adaptive Dynamic Programming algorithms HDP, DHP and GDHP.

8 pages, 4 figures. In Proceedings of the IEEE International Joint Conference on Neural Networks, June 2012, Brisbane (IEEE IJCNN 2012), pp. 3070--3077

References in corpus (3)

• Stable Adaptive Control Using New Critic Designs
• Reinforcement Learning by Value Gradients
• The Local Optimality of Reinforcement Learning by Value Gradients, and its Relationship to Policy Gradient Learning

Cited by in corpus (1)

• Learning Representations and Agents for Information Retrieval