Artificial Neural Networks: Integral Equations: Physics Informed Neural Networks

EasyChair Preprint no. 11381

Abstract

In  this  research  paper,  a  novel  model of  artificial  neuron ( in  the  spirit  of spiking  neuron ) is  summarized.  It  is  discussed,  how  integral  equations  naturally  arise  in  associative  memory based  on  such  a  model  of  artificial  neuron. This  research  paper  mainly  proposes,  the  utilization  of  Physics  Informed  Neural  Networks   ( PINNs )  for  solving  integral  equations.  We  expect  to  utilize  PINNs  for  data  driven  discovery  of   integral  equations.  It  is  expected  that  this  research  paper  will  lead  to  new  research  on  solving   non-linear  integro-differential  equations

Keyphrases: Artificial Neural Networks, dynamic synapse, integral equations, linear filter, Physics-informed neural networks

@Booklet{EasyChair:11381,
  author = {Rama Garimella},
  title = {Artificial  Neural  Networks: Integral Equations:  Physics  Informed  Neural  Networks},
  howpublished = {EasyChair Preprint no. 11381},

  year = {EasyChair, 2023}}