There has been a remarkable increase in understanding of natural adaptive systems in the last few years in areas like molecular biology, immunology, embryology, neuroscience, ecology, cognitive science, paleontology, economics, and evolution. These have important implications for artificial intelligence. In my view the main task of artificial intelligence is to produce an intelligence in the laboratory that can learn. Our largest computing problems are too complex and poorly understood for us to have any hope of simply programming solutions to them as we did in the past.
My current work is in genetic algorithms, a branch of machine learning, which is a branch of artificial intelligence.
My work focuses on the theoretical and engineering consequences of various implementations of genetic algorithms. So far my work has been restricted to proving theoretical bounds of genetic algorithm performance, and on extending the basic algorithm to more complex genetic algorithms. My future work will focus on describing just what mathematical properties of search spaces a genetic algorithm exploits during its search.
Bold student names indicate a cognitive science standalone student.
Author | Dissertation Title | Committee |
Baray, C. | Evolution of Coordination in Reactive Multi-Agent Systems (December 1999) | Mills, J. (Chair), Gasser, M., Rawlins, G., Timberlake, W. |
McGraw, G. E. Jr. | Letter Spirit (Part One): Emergent High-Level Perception of Letters Using Fluid Concepts (September 1995) | Hofstadter, D. R. (Chair), Gasser, M. Goldstone, R., Port, R. F., Rawlins, G. J. E. |
Wang, P. | Non-Axiomatic Reasoning System - Exploring the Essence of Intelligence (August 1995) | Hofstadter, D. (Chair), Townsend, J. T., Rawlins, G. J. E., Leake, D. B. |
Scherle, Ryan | Looking for a Haystack: Selecting Data Sources in a Distributed Retrieval System (November 2006) | Leake, D. (Co-Chair), Gasser, M. (Co-Chair), Mostafa, J., Rawlins, G. |