Why Experimentation can be better than `Perfect Guidance'
- Tobias Scheffer
- Russ Greiner, Dept of Computing Science; PI of AICML
- Christian Darken
Many problems correspond to the classical control task of determining the appropriate control action to take, given some (sequence of) observations. One standard approach to learning these control rules, called behavior cloning, involves watching a perfect operator operate a plant, and then trying to emulate its behavior. In the experimental learning approach, by contrast, the learner first guesses an initial operation-to-action policy and tries it out. If this policy performs sub-optimally, the learner can modify it to produce a new policy, and recur. This paper discusses the relative effectiveness of these two approaches, especially in the presence of perceptual aliasing, showing in particular that the experimental learner can often learn more effectively than the cloning one.
Citation
T. Scheffer, R. Greiner, C. Darken. "Why Experimentation can be better than `Perfect Guidance'". International Conference on Machine Learning (ICML), Nashville, July 1997.| Keywords: | guidance, experimentation, machine learning, PAC learning |
| Category: | In Conference |
BibTeX
@incollection{Scheffer+al:ICML97,
author = {Tobias Scheffer and Russ Greiner and Christian Darken},
title = {Why Experimentation can be better than `Perfect Guidance'},
booktitle = {International Conference on Machine Learning (ICML)},
year = 1997,
}Last Updated: April 25, 2007Submitted by AICML Admin Assistant
