多项式感知器用于求解胡尔维茨可满足问题

Polynomial Perceptron for H-Polynomial

定义了多项式感知器 ,分析了多项式感知器的逼近能力及特性 ,给出了多项式感知器用于求解多项式胡尔维茨可满足问题的神经网络学习算法。算例表明 ,该算法简单可行 ,易于实现 ,在稳定性问题中有着广泛的应用前景。

The concepts of polynomial perceptron are proposed. The nonlinear ability of polynomial perceptron is analyzed, and a neural network learning algorithm of polynomial perceptron is applied to solve H-polynomial. An example illustrates its efficiency, and the algorithm has very important practical significance in stability problem.