Stochastic optimal control and sensori-motor integration

The paper discusses the problem of modelling intelligent behaviour using stochastic optimal control theory. The stochastic control solution requires state feed-back which requires vast computational resources both in terms of memory and computation. We argue that an efficient approach to this problem requires an integration of sensory and motor computation. We propose the path integral control framework as a natural theory for sensori-motor integration using a Bayesian framework.

[1]  Emanuel Todorov,et al.  Linearly-solvable Markov decision problems , 2006, NIPS.

[2]  Emanuel Todorov,et al.  Efficient computation of optimal actions , 2009, Proceedings of the National Academy of Sciences.

[3]  H. Kappen.  Linear theory for control of nonlinear stochastic systems. , 2004, Physical review letters.

[4]  Raymond J. Dolan,et al.  Game Theory of Mind , 2008, PLoS Comput. Biol..

[5]  Stefan Schaal,et al.  Reinforcement learning of motor skills in high dimensions: A path integral approach , 2010, 2010 IEEE International Conference on Robotics and Automation.

[6]  Vicenç Gómez,et al.  Optimal control as a graphical model inference problem , 2009, Machine Learning.