@inproceedings {pub4049,
	title = {Learning Transferable Variation Operators in a Continuous Genetic Algorithm},
	author = {Stephen Friess AND Peter Tino AND Stefan Menzel AND Bernhard Sendhoff AND Xin Yao},
	year = {2019},
	month = {December},
	abstract = {The notion of experience has long been neglected within the domain of evolutionary computation. While in machine learning a large variety of methods has emerged in the recent years under the umbrella of transfer learning, a similar notion for experience reuse has been missing in optimization. Notably, realizing experience-based methods suffers from a variety of conceptual key problems. The first one being in regards to what constitutes problem-similarity from an algorithm perspective and the second one being what constitutes the transferable experience by itself. Ideally, one would envision that a learning optimization algorithm could be expected to act similar to a human-problem solver. Tackling a task without preconceptions only until sufficient similarity to known problems is established. Our paper therefore has two aims. First, to outline existing related fields and methodologies and highlight their insufficiencies. Second, to make the case for experience-based optimization by a demonstration using a novel and statistics-based approach with a real-coded genetic algorithm as base. In this paper we do not claim to make the case to construct universal problem solvers. But instead propose that from an algorithm-specific-view, problems reveal peculiarities which can be harnessed to improve future performance upon similar-structured optimization tasks. },
	publisher = {IEEE},
	booktitle = {IEEE Symposium Series on Computational Intelligence}
}