WP1: Unified model of evolution
The first work package is designed to provide the foundation that will allow a sustainable collaboration between population genetics and evolutionary computation, and the development of a new scientific field. Results and techniques from both fields will be transferred via a unified model and terminology. New analytical techniques for deriving results about artificial and natural evolution will be developed. Fundamental limitations to the speed of evolution will be explored.
WP2: Features of fitness landscapes and the speed of adaptation
The second work package is designed to uncover how the speed of adaptation in an asexually evolving population depends on characteristics of the fitness landscape. The results will make it easier to identify which algorithms are appropriate for a given fitness landscape. The results will serve as a foundation for the analysis of more complex evolutionary models in WP3 and WP4.
WP3: Recombination and the speed of adaptation
This work package will describe what role recombination has in evol ution, in particular with respect to the speed of adaptation, and how it interacts with other aspects of the evolutionary process, such as mutation and selection intensity. This knowledge will lead to recommendations on how to design more powerful recombination operators.
WP4: Parallel evolution in structured populations
The goal of this work package is to provide a powerful and integrated model of evolution in structured populations. It will combine existing analyses of parallel EAs with the Shifting Balance Theory (SBT) in population genetics. This novel perspective will allow us to study the speed of adaptation of EAs on combinatorial optimisation problems. The insights from these analyses will lead to recommendations on how to design effective parallel EAs.