Computer Science – Artificial Intelligence
Scientific paper
2001-11-28
Computer Science
Artificial Intelligence
This is an extended version of the paper presented at the EWRL 2001 in Utrecht (The Netherlands)
Scientific paper
We introduce a learning method called ``gradient-based reinforcement planning'' (GREP). Unlike traditional DP methods that improve their policy backwards in time, GREP is a gradient-based method that plans ahead and improves its policy before it actually acts in the environment. We derive formulas for the exact policy gradient that maximizes the expected future reward and confirm our ideas with numerical experiments.
Hutter Marcus
Kwee Ivo
Schmidhuber Juergen
No associations
LandOfFree
Gradient-based Reinforcement Planning in Policy-Search Methods 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 Gradient-based Reinforcement Planning in Policy-Search Methods, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Gradient-based Reinforcement Planning in Policy-Search Methods will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-117855