Solving complex differential equation using unsupervised neural networks

Authors

  • Christopher P. Monterola National Institute of Physics, University of the Philippines Diliman
  • Caesar A. Saloma National Institute of Physics, University of the Philippines Diliman

Abstract

The ability of neural networks (NN) to solve differential equations (DE) was demonstrated recently. In this paper, we demonstrate a general method of solving complex differential equation based on unsupervised neural networks. In particular, the paradigm is utilized to approximate the temporal profile of pulse propagating along a nonlinear and dispersive medium.

Downloads

Issue

Article ID

SPP-1998-IC-02

Section

Instrumentation and Computational Physics

Published

1998-10-27

How to Cite

[1]
CP Monterola and CA Saloma, Solving complex differential equation using unsupervised neural networks, Proceedings of the Samahang Pisika ng Pilipinas 16, SPP-1998-IC-02 (1998). URL: https://proceedings.spp-online.org/article/view/SPP-1998-IC-02.