A Framework for Multi-Paradigmatic Learning
Abstract
This paper describes our initial implementation of a domain-independent Integrated Learning System (ILS), and one application, which, through its own experience, discovers how to control a telecommunications network. ILS provides a framework for integrating several heterogeneous learning agents, in this case implementations of inductive, search-based and knowledge-based learning. These agents, written in various languages and executing on various platforms, cooperate to improve problem-solving performance. ILS also includes a central controller, called The Learning Coordinator (TLC), which manages control flow and communication between the agents using a high-level communication protocol. The agents provide TLC with expert advice. TLC chooses which suggestion to adopt and performs the appropriate actions. At intervals, the agents can inspect the results of the TLC'S actions and use this feedback to learn, improving the value of their future advice. At present ILS is being extensively tested, and the initial results are promising.
Cite
Text
Silver et al. "A Framework for Multi-Paradigmatic Learning." International Conference on Machine Learning, 1990. doi:10.1016/B978-1-55860-141-3.50045-6Markdown
[Silver et al. "A Framework for Multi-Paradigmatic Learning." International Conference on Machine Learning, 1990.](https://mlanthology.org/icml/1990/silver1990icml-framework/) doi:10.1016/B978-1-55860-141-3.50045-6BibTeX
@inproceedings{silver1990icml-framework,
title = {{A Framework for Multi-Paradigmatic Learning}},
author = {Silver, Bernard and Frawley, William J. and Iba, Glenn A. and Vittal, John and Bradford, Kelly},
booktitle = {International Conference on Machine Learning},
year = {1990},
pages = {348-356},
doi = {10.1016/B978-1-55860-141-3.50045-6},
url = {https://mlanthology.org/icml/1990/silver1990icml-framework/}
}