When Evolution is Revolution

Friday, March 24, 3pm, MC 320

James P. Crutchfield

Research Professor

Santa Fe Institute

1399 Hyde Park Road

Santa Fe, New Mexico 87501

chaos@santafe.edu

http://www.santafe.edu/~jpc

Epochal dynamics, in which long periods of stasis in an evolving population are punctuated by a sudden burst of change, is a common behavior in both natural and artificial evolutionary processes. I will review recent approaches to analyzing patterns of innovation and stasis in evolutionary search and optimization algorithms, with particular emphasis on a new mechanism that drives evolution to increased structural complexity via a sequence of phase transitions. The analysis serves to highlight the key role of non-adaptive evolution---what we now call neutrality. It also allows us to predict the optimal setting of evolutionary parameters---such as, population size and mutation rate---for minimal search time. The simplicity of the underlying mechanism suggests that epochal dynamics should be found in systems far beyond evolutionary search. With this in mind, I will conclude with comments on consequences of ignoring neutrality when viewing these systems only in terms of optimization principles and ``landscape'' metaphors.