Solving the Real Roots of Polynomials using Neural Networks

T.-M. Lin (Taiwan)


Linear Networks, Real Roots, Neural Network


In this paper, we propose a linear neural network to find the solution of polynomial with one variable. In essence, this network model is somewhat similar to the neural network based factorization model. However, it is a specific model with superior computing properties, which can generate all real roots corresponding to the polynomial with one variable. In addition, a preprocessing method to separate different real roots is also presented in this paper. Finally, several computer simulation results support our claims.

Important Links:

Go Back