Campus Event Calendar
Event Entry
New for: D3
What and Who
Computational Complexity of Parallel Evolutionary Algorithms
Dirk Sudholt
University of Birmingham
Talk
AG 1, AG 3, AG 5, SWS, AG 4, RG1, MMCI
AG Audience
English
Note: We use this to send email in the morning.
Date, Time and Location
Tuesday, 23 November 2010
10:30
30 Minutes
E1 4
3rd floor rotunda
Saarbrücken
Abstract
Bio-inspired metaheuristics like evolutionary algorithms are often applied to highly complex problems where good solutions must be obtained in short time. Parallelization is a popular way for speeding up these optimization tasks. Parallel evolutionary algorithms distribute the evolution on subpopulations, each evolving on a distinct processor or "island." Good individuals are exchanged between the islands with respect to a given communication topology connecting the islands. I will review recent theoretical advances on the computational complexity or running time of parallel evolutionary algorithms, with a focus on possible speedups in terms of the parallel computation time.
Contact
Frank Neumann
--email hidden
passcode not visible
logged in users only
Frank Neumann, 11/22/2010 16:52
Frank Neumann, 11/22/2010 09:20 -- Created document.
Frank Neumann, 11/22/2010 09:20 -- Created document.