Evolvability and Redundancy in Shared Grammar Evolution

Luerssen, Martin H.; Powers, David

doi:10.1109/cec.2007.4424495

Cited by 4 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This includes productions created for unfit solution candidates, so even a short delay leads to a substantial build-up. An earlier paper [41] provided an extensive study of this: most of the possible performance gain at the population size of 20 is observed for a single generation delay. We therefore also report on such a configuration for our problem tasks below.…”

Section: Promote Redundancymentioning

confidence: 94%

Evolving encapsulated programs as shared grammars

Luerssen

Powers

2008

Genet Program Evolvable Mach

Self Cite

View full text Add to dashboard Cite

Facilitating the discovery and reuse of modular building blocks is generally regarded as the key to achieving better scalability in genetic programming (GP). A precedent for this exists in biology, where complex designs are the product of developmental processes that can also be abstractly modeled as generative grammars. We introduce shared grammar evolution (SGE), which aligns grammatical development with the common application of grammars in GP as a means of establishing declarative bias. Programs are derived from and represented by a global context-free grammar that is transformed and extended according to another, userdefined grammar. Grammatical productions and the subroutines they encapsulate are shared between programs, which enables their reuse without reevaluation and can significantly reduce total evaluation time for large programs and populations. Several variants of SGE employing different strategies for controlling solution size and diversity are tested on classic GP problems. Results compare favorably against GP and newer techniques, with the best results obtained by promoting diversity between derived programs.

show abstract

Section: Promote Redundancymentioning

confidence: 94%

Evolving encapsulated programs as shared grammars

Luerssen

Powers

2008

Genet Program Evolvable Mach

Self Cite

View full text Add to dashboard Cite

show abstract

“…The general approach of combining a grammar with an evolutionary algorithm leads to a developmental system (Luerssen and Powers 2007). Developmental systems, and more generally genetic programming methods (Riolo et al 2007), have been used widely for the design of electrical circuits (Koza et al 2008), a problem that shares some characteristics with the design of heat exchanger networks.…”

Section: Discussionmentioning

confidence: 99%

A rewriting grammar for heat exchanger network structure evolution with stream splitting

Fraga

2009

Engineering Optimization

View full text Add to dashboard Cite

The design of cost optimal heat exchanger networks is a difficult optimisation problem due both to the nonlinear models required and also the combinatorial size of the search space. When stream splitting is considered, the combinatorial aspects make the problem even harder. This paper describes the implementation of a two level evolutionary algorithm based on a string rewriting grammar for the evolution of the heat exchanger network structure. A biological analogue of genotypes and phenotypes is used to describe structures and specific solutions respectively. The top level algorithm evolves structures while the lower level optimises specific structures. The result is a hybrid optimisation procedure which can identify the best structures including stream splitting. Case studies from the literature are presented to demonstrate the capabilities of the novel procedure.

show abstract