Optimal Reinforcement Learning for Gaussian Systems

Details Optimal Reinforcement Learning for Gaussian Systems Optimal Reinforcement Learning for Gaussian Systems

2011-06-04

arxiv.org/abs/1106.0800v3

Statistics

Machine Learning

final pre-conference version of this NIPS 2011 paper. Once again, please note some nontrivial changes to exposition and interp

Scientific paper

The exploration-exploitation trade-off is among the central challenges of reinforcement learning. The optimal Bayesian solution is intractable in general. This paper studies to what extent analytic statements about optimal learning are possible if all beliefs are Gaussian processes. A first order approximation of learning of both loss and dynamics, for nonlinear, time-varying systems in continuous time and space, subject to a relatively weak restriction on the dynamics, is described by an infinite-dimensional partial differential equation. An approximate finite-dimensional projection gives an impression for how this result may be helpful.

Affiliated with

Also associated with

No associations

Rating

Optimal Reinforcement Learning for Gaussian Systems does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Optimal Reinforcement Learning for Gaussian Systems, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Optimal Reinforcement Learning for Gaussian Systems will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-7128