---
title: "An Analytic Learning System for Specializing Heuristics"
authors:
  - "Steven Minton"
doi: ""
url: "https://ijcai.org/proceedings/1993/127"
pdf: "https://ijcai.org/proceedings/1993/0127.pdf"
venue: "Proceedings of the Thirteenth International Joint Conference on Artificial Intelligence"
year: 1993
pages: "922-928"
track: "Machine Learning"
keywords:
---

# An Analytic Learning System for Specializing Heuristics

Steven Minton

*Proceedings of the Thirteenth International Joint Conference on Artificial Intelligence*, pages 922-928.

## Abstract

This paper describes how meta-level theories are used for analytic learning in MuLTI-TAC. MULTI-TAC operationalizes generic heuristics for constraint-satisfaction problems, in order to create programs that are tailored to specific problems. For each of its generic heuristics, MULTI-TAC has a meta-theory specifically designed for operationalizing that heuristic. We present examples of the specialization process and discuss how the theories influence the tractability of the learning process. We also describe an empirical study showing that the specialized programs produced by MULTI-TAC compare favorably to hand-coded programs.