A probabilistic reasoning-based approach to machine learning

This thesis describes a novel approach to machine learning, based on the principle of learning by reasoning. Current learning systems have significant limitations like brittleness, i.e. the deterioration of performance on a different domain or problem and lack of power required for handling real-world learning problems. The goal of my research was to develop an approach in which many of these limitations are overcome in a unified, coherent and general frame-work. This learning approach is based on techniques of reasoning, such as the discovery of the underlying principle and the recognition of the deeper basis of similarity which emulate human learning. In this thesis, the limitations of current systems are discussed, along with the arguments for using a different approach to learning. This thesis then presents a Reasoning-based Approach to machine learning the author developed, along with the details of a computer implementation. It also illustrates the approach on some learning problems not directly solvable by previous approaches.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2011

Author dc:creator

• Purswani, Krishnakumar Sukhramdas

Contributors dc:contributor

• Rendell, Larry A.

• Information Science
• Computer Science

Statement dc:rights

• Copyright 1989 Purswani, Krishnakumar Sukhramdas

Language dc:language

eng

Identifier

AAI8916298
(UMI)AAI8916298

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/19623