On the Dynamics of Real Valued Hopfield Type Neural Networks Based on Ceiling Neurons

EasyChair Preprint 15830

Abstract

In  this  research  paper,  we  introduce  Hopfield-type Neural  Networks  based on  the Ceiling Neuron  Model. Inspired  by the binary Hopfield  neural  network  and  by the  definition  of Ceiling  Neuron, we  propose an extension of  the  Hopfield  Neural  Network  from  a  binary set  of  possible  states  of  a  real  valued  neuron  to a  multistate  set.  We  investigate  the  dynamics  of  the proposed  model  in asynchronous  and synchronous  update  model.

Keyphrases: Asynchronous mode, Ceiling Neuron, Hopfield neural network, Synchronous Mode, convergence theorem

@booklet{EasyChair:15830,
  author    = {Rama Murthy Garimella and Fidelis Zanetti de Castro and Aman Singh and Jyothi Prasanna and Maha Lakshmi Bairaju and Manasa Jagannadhan and Vidya Sree Vankam},
  title     = {On the Dynamics of Real Valued Hopfield Type Neural Networks Based on Ceiling Neurons},
  howpublished = {EasyChair Preprint 15830},
  year      = {EasyChair, 2025}}