Accession Number:

ADA445048

Title:

Extended Learning on SOAR

Descriptive Note:

Final rept.

Corporate Author:

MICHIGAN UNIV ANN ARBOR

Personal Author(s):

Report Date:

2006-03-14

Pagination or Media Count:

21.0

Abstract:

The major goal of this project was to develop the science and technology for building autonomous knowledge-rich learning agents - computational systems that have significant competence for performing tasks without human intervention, but also have the ability to learn new tasks and improve almost any aspect of their behavior through learning. Our plan was to concentrate on building agents with a variety of architectural learning mechanisms, including reinforcement to capture statistical regularities, episodic to capture experiences and semantic learning to capture facts. We also planned to study their integration, including integration with Soars chunking mechanism which captures procedural knowledge. The result of this project has been to develop initial versions of episodic and semantic memory, while at the same time creating a robust implementation of reinforcement learning in Soar. By the end of this project, we had implementations of all three learning mechanisms, and each of them integrated with Soars chunking mechanism. In follow on work, we are refining episodic and semantic memory and creating a complete integration of these learning mechanisms.

Subject Categories:

  • Cybernetics

Distribution Statement:

APPROVED FOR PUBLIC RELEASE