Parallelizable asychronous by blocks algorithms for neural computing

We deal with neural computing parallel algorithms suitable for parallel processing machines and apply them to solve combinatorial optimization problems. Problems are mapped onto a spin glass model then we utilize simulated annealing and mean field theory (MFT) approximation method. It is well known that the main problem of the synchronous algorithms is to be trapped in limit cycles thus we propose an extension of the MFT approximation method of (Boisson, 1993). Though we reduced parallelism, the algorithms proposed are efficient enough to avoid the limit cycles. We obtained good results in solving our NP-hard target problem, the maximum independent set graph problem.