Expressive Languages For Evolved Programs
Over the last decade evolutionary computation techniques have begun to produce human-competitive results in several areas of science and engineering. These advances have been driven by factors including the availability of faster hardware and clusters, the influx of ideas from evolutionary biology, and the development and understanding of better algorithms and representations. In this talk I will focus on representations in "genetic programming," an evolutionary computation technique in which genetic algorithms operate on and produce executable programs. I will show how increasingly expressive representations allow for the evolution of programs with types, subroutines, macros, and evolved control structures, using the Push programming language as an example of an unusually expressive representation for evolved programs. I will briefly present recent results and ongoing projects involving genetic programming with Push, and I will offer some thoughts on the relation between evolutionary computation and other forms of machine learning.
Lee Spector is a Professor of Computer Science in the School of Cognitive Science at Hampshire College and an Adjunct Professor in the Department of Computer Science at the University of Massachusetts, Amherst. He received a B.A. in Philosophy from Oberlin College in 1984, and a Ph.D. from the Department of Computer Science at the University of Maryland in 1992. Dr. Spector teaches and conducts research in artificial intelligence, artificial life, and a variety of areas at the intersections of computer science with cognitive science, physics, evolutionary biology, and the arts. He is an NSF Distinguished Teaching Scholar, the Editor-in-Chief of the Springer journal Genetic Programming and Evolvable Machines, a member of the editorial board of the MIT Press journal Evolutionary Computation, a member of the Executive Committee of the ACM Special Interest Group on Evolutionary Computation (SIGEVO), and the author of numerous publications including the book Automatic Quantum Computer Programming: A Genetic Programming Approach.