Recombination, Parallelism and the Future of Combinatorial Evolutionary Computation. - Darrell Whitley
Abstract: Every EC researcher should look in the mirror and ask: What does Evolutionary Computation have to offer to the larger field of inexact methods for combinatorial optimization? This talk will take a new look at recombination operators, and in particular how they can be used to complement Iterated Local Search. In some cases, we can design powerful deterministic recombination operators that can be used with small dynamically emergent populations. In other cases, we need to use recombination in new, more intensive ways. This talk will also look at how search methods with different bias patterns can be exploited as parallel ensembles of solvers on multicore machines.
Darrell Whitley is a Professor of Computer Science at Colorado State University. He served as the Chair of the Governing Board of the International Society of Genetic Algorithms (ISGA) from 1993 to 1997, as the Editor-in-Chief of the journal Evolutionary Computation from 1997 to 2003 and as Co-Editor-in-Chief of the ACM Transactions on Evolutionary Learning and Optimization, 2019-2021. He was Chair of the Executive Board of ACM SIGEVO from 2007 to 2011. He was named an ACM Fellow in 2019 for his contributions to the field of genetic and evolutionary computation.