---
title: "Perturbation: A Means for Guiding Generalization"
authors:
  - "Dennis F. Kibler"
  - "Bruce W. Porter"
doi: ""
url: "https://ijcai.org/proceedings/1983/100"
pdf: "https://ijcai.org/proceedings/1983/0100.pdf"
venue: "Proceedings of the Eighth International Joint Conference on Artificial Intelligence"
year: 1983
pages: "415-417"
track: "Learning and Knowledge Acquisition"
keywords:
---

# Perturbation: A Means for Guiding Generalization

Dennis F. Kibler, Bruce W. Porter

*Proceedings of the Eighth International Joint Conference on Artificial Intelligence*, pages 415-417.

## Abstract

Learning problem solving from examples suffers from three problems. First, there is a strong dependency on the order of the presented examples. Second, each example has its own peculiarities which nust be 011ercane. Third, the size of the generalization space can be huge, even if the instance language is snall. By adding perturbation operators to the concept tree each of these problems can be alleviated. This is demonstrated in a system which learns, through interaction with a teacher, to solve simultaneous linear equations.