Title: Mobility as a Predictive Measure of Evolutionary Algorithm Effectiveness

Speaker: Monte Lunacek, CSRI PhD Student
                              
Date/Time: Wednesday, June 22, 2005, 1:00-2:00 pm

Location: Building 980, Room 24 (Sandia NM)

Brief Abstract: The global search properties of heuristic search algorithms are not well understood.  In this paper, we introduce a new metric, mobility, that quantifies the dispersion of local optima visited during a search. This allows us to explore two questions: How disperse are the local optima visited during a search? How does mobility relate to algorithm performance? We compare local search with two evolutionary algorithms, CHC and CMA-ES, on a set of non-separable, non-symmetric, multi-modal test functions. Given our mobility metric, we show that algorithms visiting more disperse local optima tend to be better optimizers.