Learning Procedures by Environment-Driven Constructive Induction

Hume, David V.

doi:10.1016/B978-1-55860-141-3.50017-1

Learning Procedures by Environment-Driven Constructive Induction

David V. Hume

ICML 1990 pp. 113-121

doi:10.1016/B978-1-55860-141-3.50017-1 /icml/1990/hume1990icml-learning/

Abstract

Induction of action sequences in a simulated robot world is described. The learner passively observes examples which are procedural action sequences where properties and/or relations change in the simulation. The observed procedures are split and trimmed to form the initial concept descriptions. Active experimentation occurs when the system repeats or tests more general descriptions of the observed examples. Generalisations are formed by constructive induction using inverse resolution and tested by execution of operational actions in the simulation. Completed execution of the test confirms success. Inability to complete a concept in the simulation causes search for successively greater generalisation on demand inorder to continue execution. Background knowledge need not be present as the system invents appropriately justified concepts as required.

ICML Semantic Scholar

Cite

Text

Hume. "Learning Procedures by Environment-Driven Constructive Induction." International Conference on Machine Learning, 1990. doi:10.1016/B978-1-55860-141-3.50017-1

Markdown

[Hume. "Learning Procedures by Environment-Driven Constructive Induction." International Conference on Machine Learning, 1990.](https://mlanthology.org/icml/1990/hume1990icml-learning/) doi:10.1016/B978-1-55860-141-3.50017-1

BibTeX

@inproceedings{hume1990icml-learning,
  title     = {{Learning Procedures by Environment-Driven Constructive Induction}},
  author    = {Hume, David V.},
  booktitle = {International Conference on Machine Learning},
  year      = {1990},
  pages     = {113-121},
  doi       = {10.1016/B978-1-55860-141-3.50017-1},
  url       = {https://mlanthology.org/icml/1990/hume1990icml-learning/}
}