Bitstream Neurons for Graph Colouring

In Shawe Taylor and Zerovnik introduced the Generalised Boltzmann Machine GBM as an extension to the Boltzmann machine that enables us to map constraint problems requiring more than two states onto a recurrent neural network In experiments were performed using the Mean Field Annealing approach to graph colouring using the Petford and Welsh algorithm as a GBM In this paper we extend the use of bitstreams from the bi polar stochastically connected Boltzmann machine to a recurrent network of stochastically connected multi state bit stream neurons MSBSN We compare the performance of the re sultant network with the results obtained using Mean Field Annealing described in and the Petford and Welsh algorithm The results obtained from simulations of the MSBSN show that there is no increase in the number of iterations taken to solve a typical randomly generated k colourable graph Clearly the digital nature of the neuron combined with its simple functionality described by Shawe Taylor Van Daalen and Zhao in could lead to a fast and inexpensive hardware implementaion