A New Wave: A Dynamic Approach to Genetic Programming

摘要: Wave is a novel form of semantic genetic programming which operates by optimising the residual errors succession short runs, and then producing cumulative solution. These runs are called periods, they have heterogeneous parameters. In this paper we leverage potential Wave's heterogeneity to simulate dynamic evolutionary environment incorporating self adaptive parameters together with an innovative approach population renewal. We conduct empirical study comparing new multiple linear regression~(MLR) as well several computation~(EC) methods including known geometric programming~(GSGP) other optimised techniques. The results our investigation show that algorithm delivers consistently equal or better performance than Standard GP (both without scaling), achieves testing fitness regression, performs significantly GSGP on five six problems studied.