Accession Number:

ADA210591

Title:

Some Results on Learning

Descriptive Note:

Technical rept.,

Corporate Author:

CARNEGIE-MELLON UNIV PITTSBURGH PA ROBOTICS INST

Personal Author(s):

• Natarajan, B. K.

Report Date:

1989-02-01

Pagination or Media Count:

29.0

Abstract:

This paper presents some formal results on learning. In particular, it concerns algorithms that learn sets and functions from examples. We seek conditions necessary and sufficient for learning over a range of probabilistic models for such algorithms. This paper concerns algorithms that learn sets and functions from examples for them. The motivation behind the study is a need to better understand the class of problems known as concept learning problems in the Artificial Intelligence literature.

Descriptors:

• *ALGORITHMS
• *LEARNING
• DOCUMENTS
• ARTIFICIAL INTELLIGENCE
• MOTIVATION
• PROBABILITY
• MODELS
• PROBLEM SOLVING

Subject Categories:

• Numerical Mathematics
• Cybernetics

Distribution Statement:

APPROVED FOR PUBLIC RELEASE