Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cai, Wenxiao, Li, Zongru, Wang, Iris, Wang, Yu-Neng, Lee, Thomas H.
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2506.12610
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916847354380288
author Cai, Wenxiao
Li, Zongru
Wang, Iris
Wang, Yu-Neng
Lee, Thomas H.
author_facet Cai, Wenxiao
Li, Zongru
Wang, Iris
Wang, Yu-Neng
Lee, Thomas H.
contents Machine learning has achieved remarkable advancements but at the cost of significant computational resources. This has created an urgent need for a novel and energy-efficient computational fabric and corresponding algorithms. CMOS Oscillator Networks (OscNet) is a brain inspired and specially designed hardware for low energy consumption. In this paper, we propose a Hopfield Network based machine learning algorithm that can be implemented on OscNet. The network is trained using forward propagation alone to learn sparsely connected weights, yet achieves an 8% improvement in accuracy compared to conventional deep learning models on MNIST dataset. OscNet v1.5 achieves competitive accuracy on MNIST and is well-suited for implementation using CMOS-compatible ring oscillator arrays with SHIL. In oscillator-based inference, we utilize only 24% of the connections used in a fully connected Hopfield network, with merely a 0.1% drop in accuracy. OscNet v1.5 relies solely on forward propagation and employs sparse connections, making it an energy-efficient machine learning pipeline designed for oscillator computing fabric. The repository for OscNet family is: https://github.com/RussRobin/OscNet .
format Preprint
id arxiv_https___arxiv_org_abs_2506_12610
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle OscNet v1.5: Energy Efficient Hopfield Network on CMOS Oscillators for Image Classification
Cai, Wenxiao
Li, Zongru
Wang, Iris
Wang, Yu-Neng
Lee, Thomas H.
Computer Vision and Pattern Recognition
Machine learning has achieved remarkable advancements but at the cost of significant computational resources. This has created an urgent need for a novel and energy-efficient computational fabric and corresponding algorithms. CMOS Oscillator Networks (OscNet) is a brain inspired and specially designed hardware for low energy consumption. In this paper, we propose a Hopfield Network based machine learning algorithm that can be implemented on OscNet. The network is trained using forward propagation alone to learn sparsely connected weights, yet achieves an 8% improvement in accuracy compared to conventional deep learning models on MNIST dataset. OscNet v1.5 achieves competitive accuracy on MNIST and is well-suited for implementation using CMOS-compatible ring oscillator arrays with SHIL. In oscillator-based inference, we utilize only 24% of the connections used in a fully connected Hopfield network, with merely a 0.1% drop in accuracy. OscNet v1.5 relies solely on forward propagation and employs sparse connections, making it an energy-efficient machine learning pipeline designed for oscillator computing fabric. The repository for OscNet family is: https://github.com/RussRobin/OscNet .
title OscNet v1.5: Energy Efficient Hopfield Network on CMOS Oscillators for Image Classification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.12610