On stochastic optimal control and reinforcement learning by approximate inference (extended abstract)

Authors:
Konrad Rawlik;Marc Toussaint;Sethu Vijayakumar
Affiliations:
School of Informatics, University of Edinburgh;Inst. für Parallele und Verteilte Systeme, Universität Stuttgart;School of Informatics, University of Edinburgh
Venue:
IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence
Year:
2013

Citing 4
Cited 0

Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
Pattern Recognition and Machine Learning (Information Science and Statistics)

Pattern Recognition and Machine Learning (Information Science and Statistics)
Natural Actor-Critic

Neurocomputing
Robot trajectory optimization using approximate inference

ICML '09 Proceedings of the 26th Annual International Conference on Machine Learning

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present a reformulation of the stochastic optimal control problem in terms of KL divergence minimisation, not only providing a unifying perspective of previous approaches in this area, but also demonstrating that the formalism leads to novel practical approaches to the control problem. Specifically, a natural relaxation of the dual formulation gives rise to exact iterative solutions to the finite and infinite horizon stochastic optimal control problem, while direct application of Bayesian inference methods yields instances of risk sensitive control.