	SSP 1994 project summary:

[EPCC home] [SSP home] [2001 projects] [2000 projects] [1999 projects] [1998 projects] [1997 projects] [1996 projects] [1995 projects] [1994 projects] [1993 projects]

Visualisation of Genetic Algorithms

This project aims to investigate techniques and develop tools to aid in visualising the behaviour and execution of genetic algorithms. Many of the projects within EPCC involve GAs. Each of these could potentially benefit from having some means of visually determining the behaviour of the system. At present, debugging and watching GAs consists of dumping text diagnostics and tables, an inherently non-interactive and limited technique.

This project investigates ways of implementing this visualisation process with respect to EPCC's Genetic Algorithm tool, RPL2. Ways must be found to examine and change data held by RPL2 while executing a Genetic Algorithm, and ensuring this parallel data is given to the Visualisation tool in a helpful way.

This project investigates ways of visualising (offline and interactively) both (a) generic techniques for visualising GAs, based on attributes such as population fitness over time, and (b) specific ways of visualising an array of different problems (including for example, the Traveling Salesman Problem), including watching complex variations within the population.

Graham Clark worked on this project.

Compressed PostScript of Graham's final report is available here (307426 bytes) .

Webpage maintained by mario@epcc.ed.ac.uk