A genetic programming ecosystem

Algorithms are needed in every aspect of parallel computing. Genetic Programming is an evolutionary technique for automating the design of algorithms through iterative steps of mutation and crossover operations on an initial population of randomly generated computer programs. This paper describes a novel parallel genetic programming (GP) system inspired by the symbiogenesis model of evolution, wherein new organisms are generated through the absorption of different life-forms in addition to the usual mutation and crossover operations. Different organisms are expressed in this GP system through multiple program representations. Two program representations considered in this paper are the procedural representation (PR) and the tree representation (TR). Populations of these representations evolve separately. Individuals in each population migrate to the other and participate in evolution via representation change algorithms. Parallelism is achieved through use of the AutoMap/AutoLink MPI library. The differences in the locality properties of the representations serve as a source of new ideas for creating the final algorithm.

[1]  Riccardo Poli,et al.  Why Ants are Hard , 1998 .

[2]  Vidroha Debroy,et al.  Genetic Programming , 1998, Lecture Notes in Computer Science.

[3]  William L. George,et al.  IMPI: Making MPI Interoperable , 2000, Journal of research of the National Institute of Standards and Technology.

[4]  John R. Koza,et al.  Building a Parallel Computer System for $18, 000 that Performs a Half Peta-Flop per Day , 1999, GECCO.

[5]  John R. Koza,et al.  Genetic programming (videotape): the movie , 1992 .

[6]  Ryszard S. Michalski,et al.  Data-Driven Constructive Induction , 1998, IEEE Intell. Syst..

[7]  Martial Michel,et al.  A generalized approach for transferring data-types with arbitrary communication libraries , 2000, Proceedings Seventh International Conference on Parallel and Distributed Systems: Workshops.

[8]  Martial Michel,et al.  Fine packet size tuning with AutoLink , 1999, Proceedings of the 1999 ICPP Workshops on Collaboration and Mobile Computing (CMC'99). Group Communications (IWGC). Internet '99 (IWI'99). Industrial Applications on Network Computing (INDAP). Multime.

[9]  Martial Michel,et al.  AutoMap and AutoLink: Tools for Communicating Complex and Dynamic Data-Structures Using MPI , 1998, CANPC.

[10]  John R. Koza,et al.  Genetic Programming II , 1992 .

[11]  John R. Koza,et al.  Genetic programming II (videotape): the next generation , 1994 .

[12]  John R. Koza,et al.  A Parallel Implementation of Genetic Programming that Achieves Super-Linear Performance , 1998, Inf. Sci..

[13]  Nils J. Nilsson,et al.  Artificial Intelligence , 1974, IFIP Congress.

[14]  John H. Holland,et al.  Distributed genetic algorithms for function optimization , 1989 .

[15]  L. Margulis.  Symbiotic Planet: A New Look At Evolution , 1998 .