In contrast to simulators for ecological processes as they are designed and implemented today, the ParalLife system which is introduced in this article can make use of problem inherent parallehsm to speed up the simulation process. We describe how the simulated environment can be distributed over massively parallel computer architectures and what can be gained by doing so. We then describe Classifier Systems (CS) as one of the decision models available for Animats in ParalLiye. A second topic of this paper is an experiment where Animats with CS improve their performance with Lamarckian Inheritance by recombining parental brain substructures.
CITATION STYLE
Bartscht, E., Engel, J., & Müller-Schloer, C. (1995). The investigation of lamarckian inheritance with classifier systems in a massively parallel simulation environment. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 929, pp. 920–933). Springer Verlag. https://doi.org/10.1007/3-540-59496-5_353
Mendeley helps you to discover research relevant for your work.