About Hyperspace Graph Paper

[HSGP Example Image]
Example Hypergraph, for sHDF, a simplified version of Goldberg's HDF.

Hyperspace graph paper (HSGP) is a technique for visualising cost surfaces that are relavent to work in evolutionary computation, particularly genetic algorithms. The focus is on moderate-dimensional, binary cost surfaces (functions defined over {0, 1}n, n ≤ 16) that have a hierarchical modular structure. The visualisation approach involves an unfolding of the hyperspace into a two-dimensional graph, whose layout represents the topology of the space using a recursive relationship, and whose shading defines the shape of the cost surface defined on the space.

The technique is described fully in:

  • Wiles, J. & Tonkes, B. (2002). Visualisation of hierarchical cost surfaces for evolutionary computation. In Proceedings of the 2002 Congress on Evolutionary Computation, 157 - 162.
Available here in PostScript and PDF formats.

HSGP

If you have a Java2-capable browser with Java enabled, you can run the hyperspace graph paper tool by clicking on the button below. If you do not have a Java2-capable browser, then you can use the hyperspace graph paper tool by installing Java2 and pointing appletviewer at this page (http://www.itee.uq.edu.au/~jwiles/hgsp/index.html).


More Information

  1. Using HSGP
  2. Available Fitness Functions
  3. Download
  4. Known Bugs
  5. The Five-Minute Guide to HSGP and Hopfield Networks

  • Wiles, J. & Tonkes, B. (2002). Visualisation of hierarchical cost surfaces for evolutionary computation. In Proceedings of the 2002 Congress on Evolutionary Computation, 157 - 162.
Available here in PostScript and PDF formats.