- author = "Tina Yu",
- title = "Program Evolvability Under Environmental Variations and Neutrality",
- booktitle = "Proceedings 9th European Conference on Artificial Life, ECAL 2007",
- year = "2007",
- editor = "Fernando {Almeida e Costa} and Luis Mateus Rocha and Ernesto Costa and Inman Harvey and Antonio Coutinho",
- volume = "4648",
- series = "Lecture Notes in Computer Science",
- pages = "835--844",
- address = "Lisbon",
- month = sep # " 10-14",
- publisher = "Springer",
- keywords = "genetic algorithms, genetic programming",
- isbn13 = "978-3-540-74913-4",
-
URL = "
http://www.cs.mun.ca/~tinayu/Publications_files/ECAL2007.pdf",
-
DOI = "
doi:10.1007/978-3-540-74913-4_84",
- size = "10 pages",
- abstract = "Biological organisms employ various mechanisms to cope with the dynamic environments they live in. One recent research reported that depending on the rates of environmental variation, populations evolve toward genotypes in different regions of the neutral networks to adapt to the changes. Inspired by that work, we used a genetic programming system to study the evolution of computer programs under environmental variation. Similar to biological evolution, the genetic programming populations exploit neutrality to cope with environmental fluctuations and evolve evolvability. We hope this work sheds new light on the design of open-ended evolutionary systems which are able to provide consistent evolvability under variable conditions.",
-
notes = "EQ only function set \cite{langdon:1998:BBparity} even
4-bit parity v. 4 bit always on.
p836 'variation is the fuel of evolution'. p387 at the edge of neutral 'network mutations are likely to produce different phenotype'. No crossover. No length changes? Solution is exactly twice as likely to be selected as non-solution. I.e. always-on is no worse than anything else when selecting for even-4-parity. p842 longer solutions have proportionally more mutations. p843 ???All better programs have the maximum size (18) ???",
