Benchmarking a BI-population CMA-ES on the BBOB-2009 function testbed

We propose a multistart CMA-ES with equal budgets for two interlaced restart strategies, one with an increasing population size and one with varying small population sizes. This BI-population CMA-ES is benchmarked on the BBOB2009 noiseless function testbed and could solve 23, 22 and 20 functions out of 24 in search space dimensions 10, 20 and 40, respectively, within a budget of less than 106 D function evaluations per trial. Categories and Subject Descriptors

• Type: Conference
• Year: 2009
• Where: GECCO
• Authors: Nikolaus Hansen
• Keywords: GECCO 2009, Interlaced Restart Strategies, Optimization, Population Sizes, Small Population Sizes