Abstract
This paper describes a method for evolutionary program induction of binary Java bytecode. Like many other machine code based methods it uses a linear genome. The genetic operators are adapted to the stack architecture and preserve stack depth during crossover. In this work we have extended a previous system to run in a distributed manner on several different physical machines. We call our new system Distributed Java Bytecode Genetic Programming (DJBGP). We use the Voyager package for migration of Java individuals. The system’s feasibility is demonstrated on a telecom routing problem.
Cite
CITATION STYLE
Lukschandl, E., Borgvall, H., Nohle, L., Nordahl, M., & Nordin, P. (2000). Distributed java bytecode genetic programming with telecom applications. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1802, pp. 316–325). Springer Verlag. https://doi.org/10.1007/978-3-540-46239-2_24
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
