Scaling Equilibrium Propagation to Deeper Neural Network Architectures

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Elayedam, Sankar Vinayak, Srinivasan, Gopalakrishnan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914403599777792
author Elayedam, Sankar Vinayak
Srinivasan, Gopalakrishnan
author_facet Elayedam, Sankar Vinayak
Srinivasan, Gopalakrishnan
contents Equilibrium propagation has been proposed as a biologically plausible alternative to the backpropagation algorithm. The local nature of gradient computations, combined with the use of convergent RNNs to reach equilibrium states, make this approach well-suited for implementation on neuromorphic hardware. However, previous studies on equilibrium propagation have been restricted to networks containing only dense layers or relatively small architectures with a few convolutional layers followed by a final dense layer. These networks have a significant gap in accuracy compared to similarly sized feedforward networks trained with backpropagation. In this work, we introduce the Hopfield-Resnet architecture, which incorporates residual (or skip) connections in Hopfield networks with clipped $\mathrm{ReLU}$ as the activation function. The proposed architectural enhancements enable the training of networks with nearly twice the number of layers reported in prior works. For example, Hopfield-Resnet13 achieves 93.92\% accuracy on CIFAR-10, which is $\approx$3.5\% higher than the previous best result and comparable to that provided by Resnet13 trained using backpropagation.
format Preprint
id arxiv_https___arxiv_org_abs_2509_26003
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Scaling Equilibrium Propagation to Deeper Neural Network Architectures
Elayedam, Sankar Vinayak
Srinivasan, Gopalakrishnan
Neural and Evolutionary Computing
Machine Learning
Equilibrium propagation has been proposed as a biologically plausible alternative to the backpropagation algorithm. The local nature of gradient computations, combined with the use of convergent RNNs to reach equilibrium states, make this approach well-suited for implementation on neuromorphic hardware. However, previous studies on equilibrium propagation have been restricted to networks containing only dense layers or relatively small architectures with a few convolutional layers followed by a final dense layer. These networks have a significant gap in accuracy compared to similarly sized feedforward networks trained with backpropagation. In this work, we introduce the Hopfield-Resnet architecture, which incorporates residual (or skip) connections in Hopfield networks with clipped $\mathrm{ReLU}$ as the activation function. The proposed architectural enhancements enable the training of networks with nearly twice the number of layers reported in prior works. For example, Hopfield-Resnet13 achieves 93.92\% accuracy on CIFAR-10, which is $\approx$3.5\% higher than the previous best result and comparable to that provided by Resnet13 trained using backpropagation.
title Scaling Equilibrium Propagation to Deeper Neural Network Architectures
topic Neural and Evolutionary Computing
Machine Learning
url https://arxiv.org/abs/2509.26003