---
title: "Verification-based Learning: A Generalized Strategy for Inferring Problem-Reduction Methods"
authors:
  - "Sridhar Mahadevan"
doi: ""
url: "https://ijcai.org/proceedings/1985/119"
pdf: "https://ijcai.org/proceedings/1985/0119.pdf"
venue: "Proceedings of the Ninth International Joint Conference on Artificial Intelligence"
year: 1985
pages: "616-622"
track: "Learning and Acquisition"
keywords:
---

# Verification-based Learning: A Generalized Strategy for Inferring Problem-Reduction Methods

Sridhar Mahadevan

*Proceedings of the Ninth International Joint Conference on Artificial Intelligence*, pages 616-622.

## Abstract

A major impediment to the develo pment of high-performance knowledge-based systems arises from the prohibitive effort involved in equipping these systems with a sufficient set of problem-solving methods . Thus , one imp o rtant research problem in Machine Learning has been the st udy of techniques for inferring problem-solving methods from e xampl es. Although a number of techniques for learning problem-s olving methods have been described in th e literature , all of them assume a .tate-apace model of problem-solving. In this paper we describe a new te chnique for learning problem-reduction meth ods, Verification-Based Learning (VBL) , which extends the earlier techniques to the problem-reduction formulation of pr oblem- solving . We illustrate the VBL technique with examples drawn from circuit design and symbolic integration.