To optimize unknown 'fitness' functions, we present Natural Evolution Strategies, a novel algorithm that constitutes a principled alternative to standard stochastic search methods. It maintains a multinormal distribution on the set of solution candidates. The Natural Gradient is used to update the distribution's parameters in the direction of higher expected tness, by eficiently calculating the inverse of the exact Fisher information matrix whereas previous methods had to use approximations. Other novel aspects of our method include optimal tness baselines and importance mixing, a procedure adjusting batches with minimal numbers of tness evaluations. The algorithm yields competitive results on a number of benchmarks. Copyright 2009.
|Original language||English (US)|
|Title of host publication||ACM International Conference Proceeding Series|
|State||Published - Sep 15 2009|