A depth controlling strategy for Strongly Typed Evolutionary Programming
Claire J. Kennedy, Christophe G. Giraud-Carrier · 1999
This paper presents a dynamic strategy for monitoring the depth of program trees evolved by STEPS (Strongly Typed Evolutionary Programming System). STEPS evolves higher-order functional programs in the form of trees, which are allowed to grow or shrink to fit the size of the problem, via specialised genetic operators. Thus, the need for arbitrary cut-off mechanisms is eliminated. 1 INTRODUCTION Most evolutionary algorithms rely on fixed-length representations. Clearly, such representations simplify implementations. However, they often require the user to have some knowledge of the appearance and structure of the final solution. More recently, variable-length representations have been used to alleviate these limitations. One notable example is in the area of Genetic Programming where programs in the form of parse trees are evolved (Koza 1992). Although, more flexible and less demanding of prior knowledge, the variable- length representation can lead to a general increase in the depth ...