IJCSNS International Journal of Computer Science and Network Security, VOL.9 No.4, April 2009
Manuscript received April 5, 2009
Manuscript revised April 20, 2009
A New Algorithm Using Hopfield Neural Network with CHN for
Wei Zhang† and Zheng Tang††,
Faculty of Engineering, Toyama University, Toyama-Shi, JAPAN
A model of neurons with CHN (Continuous Hysteresis Neurons)
for the Hopfield neural networks is studied. We prove
theoretically that the emergent collective properties of the
original Hopfield neural networks also are present in the
Hopfield neural networks with continuous hysteresis neurons.
The network architecture is applied to the N-Queens problem and
results of computer simulations are presented and used to
illustrate the computation power of the network architecture. The
simulation results show that the Hopfield neural network with
CHN is much better than other algorithms for N-Queens problem
in terms of both the computation time and the solution quality.
Hopfield neural network, hysteresis, collective properties, N-
The optimization problems are encountered in various
situation. There is a problem which has discrete
Optimization Problems.” This problem is complicated
more than a linear programming, and it is called “NP-
Hard.” N-Queens problem is one of the NP-Hard
combinatorial optimization problems. N-Queens problem
is that N chess queens must be placed on a square
chessboard composed of N rows and N columns, in such a
way that they do not attack each other for 8 directions.
The auto associative memory model proposed by Hopfield
[1, 2] has attracted considerable interest both as a content
address memory (CAM) and, more interestingly, as a
method of solving difficult optimization problems [3-5].
interconnected nonlinear processing elements (“neurons”)
with two-state threshold neurons