—This paper proposes a quantum competition neural network model compared to its classical counterpart from the relative parts of the complex system localizing operation without changing the perspective of entanglement measure. It shows that the pseudo-state is an inevitable part of the quantum competitive model. After the initialization of the quantum neural network; quantum competitive network is capable of associative memory through local area of operations because of the existence of these pseudo-states. Furthermore, Competitive algorithms of quantum theory are given, and finally an example of pattern recognition for simulation. Simulation results show that a quantum competitive learning algorithm in the learning rate and convergence rate is far better than the basic competitive artificial neural network.
展开▼
