Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
409438 | Neurocomputing | 2006 | 5 Pages |
Abstract
In this paper, we proposed an improved discrete Hopfield neural network (DHNN) for Max-Cut problems. By introducing a nonlinear self-feedback term to the motion equation of the DHNN, the DHNN can escape from local minima and therefore get better solutions. Simulation results show that the proposed algorithm has superior ability for Max-Cut problems within reasonable number of iterations.
Related Topics
Physical Sciences and Engineering
Computer Science
Artificial Intelligence
Authors
Jiahai Wang,