ISSN:
1573-1995
Source:
Springer Online Journal Archives 1860-2000
Topics:
Electrical Engineering, Measurement and Control Technology
Notes:
Abstract The Eos platform supports research and rapid implementation of evolutionary algorithms, ecosystem simulations and hybrid models. It also supports fast prototyping of industrial applications using these technologies. A large and rapidly growing library of evolutionary algorithm types and options is provided which together with a flexible configuration system allows a 'plug-and-play' construction of novel algorithms. Support for ecosystem models includes classes for multiple types of physical space (n-dimensional discrete or continuous Cartesian space, graph space), complex interactions between entities, and movement of individuals between populations. The flexibility of the Eos platform is expected to provide a powerful environment for developing new algorithms and architectures. Eos is implemented in Java™ for portability and to allow easy extension of the core functionality. It supports transparent distribution of evolutionary and ecosystem implementations across multi-processor computer clusters. This paper describes the architecture and functionality of the Eos platform and illustrates its use by way of a number of example applications.
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1023/A:1026798423593
