Short term memory in genetic programming

Created by W.Langdon from gp-bibliography.bib Revision:1.8120

@InProceedings{Bearpark:2000:ACDM,

• author = "K. Bearpark and A. J. Keane",
• title = "Short term memory in genetic programming",
• booktitle = "Fourth International Conference on Adaptive Computing in Design and Manufacture, ACDM '00",
• year = "2000",
• editor = "I. C. Parmee",
• pages = "309--320",
• address = "University of Plymouth, Devon, UK",
• publisher = "Springer-Verlag",
• keywords = "genetic algorithms, genetic programming",
• isbn13 = "978-1-85233-300-3",
• URL = "http://eprints.soton.ac.uk/21399/1/bear_00.pdf",
• URL = "http://eprints.soton.ac.uk/21399/",
• URL = "http://www.springer.com/engineering/mechanical+engineering/book/978-1-85233-300-3",
• URL = "http://www.amazon.co.uk/Evolutionary-Design-Manufacture-Selected-Papers/dp/1852333006",
• DOI = "doi:10.1007/978-1-4471-0519-0_25",
• size = "12 pages",
• abstract = "The recognition of useful information, its retention in memory, and subsequent use plays an important part in the behaviour of many biological species. Information gained by experience in one generation can be propagated to subsequent generations by some form of teaching. Each generation can then supplement its taught learning by its own experience. In this paper we explore the role of memorised information in the performance of a Genetic Programming (GP) system that uses a tree structure as its representation. Memory is implemented in the form of a set of subtrees derived from successful members of each generation. The memory is used by a genetic operator similar to the mutation operator but with the following difference. In a tree-structured system the mutation operator replaces randomly selected sub-trees by new randomly-generated sub-trees. The memory operator replaces randomly selected sub-trees by sub-trees randomly randomly selected from the memory. To study the memory operator's impact a GP system is used to evolve a well-known expression from classical kinetics using fitness-based selection. The memory operator is used together with the common crossover and mutation operators. It is shown that the addition of a memory operator increases the probability of a successful evolution for this particular problem. At this stage we make no claim for its impact on other problems that have been successfully addressed by Genetic Programming",
• notes = "Evolutionary Design and Manufacture: Selected Papers from . (ACDM '00) One example physics integration of u*t+0.5*a*t*t t=1...10, u=20 or u=200 a=980 Reverse Polish RPN except for first (in Lisp) max length=11?? 19??, roulette wheel, crossover, mutation. Memory operator: when fitness improves over best of previous generation whole of tree and its subtrees are saved in memory. Later random choices from memory. elitism.Pop=2000, gen=20 40000 tests per minute (300 MHz).",

}

Genetic Programming entries for Keith Bearpark Andy J Keane

Citations