Recently there are great research efforts devoted to address the aforementioned embarrassing issue. This book presents a series of efforts in this direction.  It is composed of four parts. Part I briefly introduces evolutionary learning and some preliminaries. Part II presents general theoretical tools for the analysis of running time and approximation performance of evolutionary algorithms. Based on these general tools, Part III presents a series of theoretical results about major factors of evolutionary optimization, such as recombination, representation, inaccurate evaluation, population, noise. Part IV presents development of evolutionary learning algorithms with provable theoretical guarantees for some representative tasks, where evolutionary learning offers excellent performance. weiterlesen