Investigating the impact of adaptation sampling in natural evolution strategies on black-box optimization testbeds

Natural Evolution Strategies (NES) are a recent member of the class of real-valued optimization algorithms that are based on adapting search distributions. Exponential NES (xNES) are the most common instantiation of NES, and particularly appropriate for the BBOB 2012 benchmarks, given that many are non-separable, and their relatively small problem dimensions. The technique of adaptation sampling, which adapts learning rates online further improves the algorithm's performance. This report provides an extensive empirical comparison to study the impact of adaptation sampling in xNES, both on the noise-free and noisy BBOB testbeds.

[1]  Nikolaus Hansen,et al.  Completely Derandomized Self-Adaptation in Evolution Strategies , 2001, Evolutionary Computation.

[2]  Shun-ichi Amari,et al.  Natural Gradient Works Efficiently in Learning , 1998, Neural Computation.

[3]  Tom Schaul,et al.  Stochastic search using the natural gradient , 2009, ICML '09.

[4]  Tom Schaul,et al.  Natural Evolution Strategies , 2008, 2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence).

[5]  Raymond Ros,et al.  Real-Parameter Black-Box Optimization Benchmarking 2009: Experimental Setup , 2009 .

[6]  Isao Ono,et al.  Proposal of distance-weighted exponential natural evolution strategies , 2011, 2011 IEEE Congress of Evolutionary Computation (CEC).

[7]  Tom Schaul,et al.  Natural evolution strategies converge on sphere functions , 2012, GECCO '12.

[8]  Anne Auger,et al.  Real-Parameter Black-Box Optimization Benchmarking 2009: Noiseless Functions Definitions , 2009 .

[9]  K. Price.  Differential evolution vs. the functions of the 2/sup nd/ ICEO , 1997, Proceedings of 1997 IEEE International Conference on Evolutionary Computation (ICEC '97).

[10]  Tom Schaul,et al.  Benchmarking exponential natural evolution strategies on the noiseless and noisy black-box optimization testbeds , 2012, GECCO '12.

[11]  Tom Schaul,et al.  Studies in Continuous Black-box Optimization , 2011 .

[12]  Tom Schaul,et al.  Benchmarking natural evolution strategies with adaptation sampling on the noiseless and noisy black-box optimization testbeds , 2012, GECCO '12.

[13]  Tom Schaul,et al.  A Natural Evolution Strategy for Multi-objective Optimization , 2010, PPSN.

[14]  Tom Schaul,et al.  Exponential natural evolution strategies , 2010, GECCO '10.