用振荡器网络实现低功耗图像识别，仅靠前向传播就提升准确率。
OscNet v1.5: Energy Efficient Hopfield Network on CMOS Oscillators for Image Classification
- 基于振荡器的霍普菲尔德网络，仅用前向传播训练。
- MNIST上比传统模型高8%准确率，连接数减少76%。
- 适合振荡器硬件，能效高，适合边缘计算场景。
机器学习虽进展迅速，但消耗大量算力。为此亟需新型节能计算架构与算法。本文提出一种可在振荡器网络（OscNet）上实现的霍普菲尔德网络算法。该网络仅通过前向传播训练，学习稀疏连接权重，在MNIST数据集上较传统深度学习模型提升8%准确率。OscNet v1.5在MNIST上表现良好，可利用CMOS兼容的环形振荡器阵列结合SHIL实现。在振荡器推理中，仅使用全连接霍普菲尔德网络24%的连接，准确率仅下降0.1%。该方案完全依赖前向传播，采用稀疏连接，专为振荡器计算硬件设计，具备极低能耗优势。
原文摘要 · Abstract (English)
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 .
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。