D-CIS Publication Database

Publication

Type of publication:Inproceedings
Entered by:BvB
TitleA path integral approach to agent planning
Bibtex cite ID
Booktitle Proceedings of the NIPS 2006 Workshop ‘Towards a New Reinforcement Learning?’
Year published 2006
Month December
Location 8 December 2006, Whistler, BC, Canada
Keywords stochastic optimal control,path integral
Abstract
Control theory is a mathematical description of how to act optimally to gain future rewards. In this paper We discuss a class of non-linear stochastic control problems that can be efficiently solved using a path integral. In this control formalism, the central concept of cost-to-go or value function becomes a free energy and methods and concepts from statistical physics can be readily applied, such as Monte Carlo sampling or the Laplace approximation. When applied to a receding horizon problem in a stationary environment, the solution resembles the one obtained by traditional reinforcement learning with discounted reward. It is shown that this solution can be computed more efficiently than in the discounted reward framework. As shown in previous work, the approach is easily generalized to time-dependent tasks and is therefore of great relevance for modeling real-time interactions between agents.
Authors
Kappen, Hilbert J.
Wiegerinck, Wim
van den Broek, Bart
Topics
=SEE CLASSIFICATION DIFFERENCE FROM OTHERS=
BibTeXBibTeX
RISRIS
Attachments
 
Total mark: 5