Versatile Graph Planarisation via an Artificial Neural Network
An artificial neural network which is based on harmony theory has been applied to the graph planarisation problem. The presented artificial neural network is capable of solving both aspects of graph planarisation (creation of the optimally planarised as well as of the maximally planar forms of any graph). Optimal solutions are always produced independent of the complexity or planarity of the original graph; if more than one equally good solutions exist they are produced with the same probability.
KeywordsArtificial Neural Network Simulated Annealing Lower Layer Graph Planarisation Uniqueness Constraint
Unable to display preview. Download preview PDF.
- P. Smolensky. Information processing in dynamical systems: foundations of harmony theory, volume 1, pages 184–281. MIT Press, Cambridge MA, 1986.Google Scholar