Tower of Hanoi with Connectionist Networks: Learning New Features

Abstract

A connectionist system previously used to solve the numerical control task of balancing a pole (Barto, Sutton, and Anderson, 1983; Anderson, 1987) is applied to a Tower of Hanoi puzzle. The connectionist system consists of two networks: an evaluation network that learns an evaluation function of states, and an action network that learns to select actions as a function of the puzzle's state and previous actions. The initial state representation is insufficient–new features must be learned to form a useful evaluation function. Comparisons of methodology are made with Langley's (1985) adaptive production system, SAGE.2.

Cite

Text

Anderson. "Tower of Hanoi with Connectionist Networks: Learning New Features." International Conference on Machine Learning, 1989. doi:10.1016/B978-1-55860-036-2.50088-6

Markdown

[Anderson. "Tower of Hanoi with Connectionist Networks: Learning New Features." International Conference on Machine Learning, 1989.](https://mlanthology.org/icml/1989/anderson1989icml-tower/) doi:10.1016/B978-1-55860-036-2.50088-6

BibTeX

@inproceedings{anderson1989icml-tower,
  title     = {{Tower of Hanoi with Connectionist Networks: Learning New Features}},
  author    = {Anderson, Charles W.},
  booktitle = {International Conference on Machine Learning},
  year      = {1989},
  pages     = {345-349},
  doi       = {10.1016/B978-1-55860-036-2.50088-6},
  url       = {https://mlanthology.org/icml/1989/anderson1989icml-tower/}
}