A parallel computation network for the maximum clique problem

The maximum clique problem is to find the maximum complete subgraph of a given graph G. A computation model for large-scale maximum clique problems is proposed and was tested. A parallel algorithm based on the maximum neural network which resembles the winner-takes-all circuit is introduced which solves large-scale problems in reasonable computation time that the best known algorithms cannot solve. The maximum clique problem is first formulated as an unconstrained quadratic zero-one programming problem and is solved by minimizing the weight summation over the same partition in a newly constructed graph.<<ETX>>