---
title: "Determining What to Learn Through Component-Task Modeling"
authors:
  - "Bruce Krulwich"
  - "Lawrence Birnbaum"
  - "Gregg Collins"
doi: ""
url: "https://ijcai.org/proceedings/1995/58"
pdf: "https://ijcai.org/proceedings/1995/0058.pdf"
venue: "Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence"
year: 1995
pages: "439-446"
track: "Cognitive Modelling"
keywords:
---

# Determining What to Learn Through Component-Task Modeling

Bruce Krulwich, Lawrence Birnbaum, Gregg Collins

*Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence*, pages 439-446.

## Abstract

Research in machine learning has typically addressed the problem .of how and when to learn, and ignored the problem of formulating learning tasks in the first place. This paper addresses this issue in the context of the CASTLE system, I that dynamically formulates learning tasks for a given situation. Our approach utilizes an explicit model of the decision-making process to pinpoint which system component should be improved. CASTLE can then focus the learning process on the issues involved in improving the performance of the particular component.