作者: Terence Soule
DOI:
关键词: Computer science 、 Voting 、 Task (project management) 、 Genetic programming 、 Knowledge management 、 Domain (software engineering) 、 Specialization (functional) 、 Function (engineering) 、 Mechanism (biology) 、 Problem domain
摘要: Cooperative teams can be used to improve the performance of standard genetic programming. However, designing appropriate team member choice and cooperation mechanisms for producing successful is an important sometimes difficult task. This task complicated by type programs being generated. For example, boolean valued outputs use simple voting as a mechanism whereas with real cannot. research examines several linear regression problem. The results suggests that produces most improvement when favors specialization. The best are achieved each focus on relatively small portion problem domain member's influence outside small. Specialization found function both mechanism.