【摘要】 提出了一种基于神经网络的多项式实根隔离算法.神经网络的权值是区间而非数值;神经网络的学习是基于区间分析的全局优化方法.该算法具有并行快速等特点,仿真实验证明了该算法的有效性.更多还原

【Abstract】 In this paper,we put forward a method for isolating real roots of polynomial based on neural network.Different from the classical neural networks,the weights of the neural network are intervals instead of numbers and the learning algorithm of the neural network is based on interval global optimization.In classical neural networks,sub-optimal solutions are frequently trapped in local minima,because they are usually based on gradient-based procedures.Our method can not only overcome these disadvantages but have the property of parallel architecture as well.Simulation experiment showed effectiveness of the method.更多还原