Front Page

Welcome to the community page for

This page is maintained by Jeff Clune (jclune theAtSign gmail dotcom). If you would like to edit this page, or have any questions, please email me.

*** Please submit papers to GECCO's 2011 Generative and Developmental Systems track, the premier conference on Generative, Indirect, and Developmental Encodings worldwide!***

This page contains information and work related to Evolutionary Algorithms that use generative encodings. Encodings are the way information is stored in a genome and the process that turns that information into a phenotype. Generative encodings reuse information in the genotype to influence many parts of the phenotype. With this technology, a small genome can encode for a larger, more complex phenotype. Generative encodings are contrasted with direct encodings, wherein each piece of information in a genotype describes a separate part of the phenotype.

This field goes by many names. Common synonyms for generative encodings are developmental or indirect encodings. Additional names for work in this field are as follows: artificial development, artificial embryogeny, computational embryology, generative systems, genetic regulatory networks (GRNs), Lindenmayer Systems (L-Systems), genotype to phenotype mappings, evolutionary design, etc. A nice review of the field and its terminology can be found in Stanley and Miikkulainen 2003.

Feel free to add content to this page, including your own work. To do so simply email me (jclune theAtSign gmail dotcom) and I will give you access.

There are many different generative encodings. Here is a quick description of a few of them, and a link to a page where you can find related publications and software.

• Hypercube-based NEAT (Stanley , D’Ambrosio & Gauci 2009)
  

• Self Modifying Cartesian Genetic Programming (SMCGP) (S. Harding, J. F. Miller, W. Banzhaf 2003)
  

• Multicellular Development with Cartesian Genetic Programming: French Flags Cartesian Genetic Programming and French Flags (J. F. Miller et al. 2003)
  

• Growth processes for modular robots, with Gene Regulatory Networks Artificial Ontogeny (J. C. Bongard et al. 2001)
  

• Defining shapes as a system-environment interaction Dynamical Blueprints (Nicolás S. Estévez, Hod Lipson 2007)
  

• Evoneuro encoding (Mouret, Doncieux & Girard 2010)
  

• Lindenmayer Systems (L-Systems)
  

• Genetic Regulatory Networks (GRNs)
  

• To be continued