Shaping in Reinforcement Learning by Changing the Physics of the Problem

Abstract

Children learn to ride a bicycle by using training wheels. They are actually trying to learn one task (riding without training wheels) by training another one. In general, solving a difficult problem can be facilitated by training other problems.

Cite

Text

Randløv. "Shaping in Reinforcement Learning by Changing the Physics of the Problem." International Conference on Machine Learning, 2000.

Markdown

[Randløv. "Shaping in Reinforcement Learning by Changing the Physics of the Problem." International Conference on Machine Learning, 2000.](https://mlanthology.org/icml/2000/randlv2000icml-shaping/)

BibTeX

@inproceedings{randlv2000icml-shaping,
  title     = {{Shaping in Reinforcement Learning by Changing the Physics of the Problem}},
  author    = {Randløv, Jette},
  booktitle = {International Conference on Machine Learning},
  year      = {2000},
  pages     = {767-774},
  url       = {https://mlanthology.org/icml/2000/randlv2000icml-shaping/}
}