- author = "James Alfred Walker and Julian Francis Miller",
- title = "The Automatic Acquisition, Evolution and Reuse of Modules in Cartesian Genetic Programming",
- journal = "IEEE Transactions on Evolutionary Computation",
- year = "2008",
- volume = "12",
- number = "4",
- pages = "397--417",
- month = aug,
- keywords = "genetic algorithms, genetic programming, Automatically defined functions (ADFs), Cartesian genetic programming (CGP), embedded Cartesian genetic programming (ECGP), genetic programming (GP), graph-based representations, modularity, module acquisition, ECGP",
- ISSN = "1089-778X",
-
DOI = "
doi:10.1109/TEVC.2007.903549",
-
URL = "
http://results.ref.ac.uk/Submissions/Output/3354578",
- size = "21 pages",
- abstract = "This paper presents a generalisation of the graph-based genetic programming (GP) technique known as Cartesian genetic programming (CGP). We have extended CGP by using automatic module acquisition, evolution, and reuse. To benchmark the new technique, we have tested it on: various digital circuit problems, two symbolic regression problems, the lawnmower problem, and the hierarchical if-and-only-if problem. The results show the new modular method evolves solutions quicker than the original nonmodular method, and the speedup is more pronounced on larger problems. Also, the new modular method performs favourably when compared with other GP methods. Analysis of the evolved modules shows they often produce recognisable functions. Prospects for further improvements to the method are discussed.",
-
notes = "8-even parity, 3-bit adder, 3-multiplier,
3-comparison, HIFF, ADF. Refers to PDGP, pushGP, ADM.
pop=5 (1+4)-ES, 1000 generations. suggestion (p401)
that ECGP may suffer bloat, stack overflow and out of
memory errors.
Combined with duplicate entry \cite{DBLP:journals/tec/WalkerM08} October 2010.
INSPEC Accession Number: 10118371",
- uk_research_excellence_2014 = "The paper advances evolutionary computing. It is the definitive journal article on automatically evolved sub-functions in Cartesian Genetic Programming (CGP). CGP, invented by Miller, has become a highly cited technique used in evolutionary computation. The paper includes detailed statistically rigorous comparisons with other GP methods and shows that CGP is one of the most efficient forms of Genetic Programming. The results contributed to work undertaken in an EPSRC funded project (EP/F062192/1).",
