An adaptive knowledge evolution strategy for finding near-optimal solutions of specific problems

摘要

Most real-world problems cannot be mathematically defined and/or structured modularly for peerresearchers in the same community to facilitate their work. This is partially because there are no concretedefined methods that can help researchers clearly describe their problems and partially because onemethod fits one problem but does not apply to others. In order to apply someone’s research results tonew domains and for researchers to collaborate with each other more efficiently, a well-defined architecturewith self-adaptive evolution strategies is proposed. It can automatically find the best solutions fromexisting knowledge and previous research experiences. The proposed architecture is based on object-orientedprogramming skills that in turn become foundations of the community interaction evolution strategyand knowledge sharing mechanism. They make up an autonomous evolution mechanism using aprogressive learning strategy and a common knowledge packaging definition. The architecture definesfourteen highly modular classes that allow users to enhance collaboration with others in the same or similarresearch community. The presented evolution strategies also integrate the merits of users’ predefinedalgorithms, group interaction and learning theory to approach the best solutions of specific problems.Finally, resource limitation problems are tackled to verify both the re-usability and flexibility of the proposedwork. Our results show that even without using any specific tuning of the problems, optimal ornear-optimal solutions are feasible.