arXiv：cs.LG· Thomas Chaffey·· 4 小时前AI 评分33
电阻二极管网络如何实现单调算子均衡网络并支持硬件线性化训练
Circuit realization and hardware linearization of monotone operator equilibrium networks
AI 导读
研究表明，电阻-二极管网络的端口行为对应 ReLU 单调算子均衡网络的解，可用模拟硬件简洁地构建这类无限深度网络。作者提出"硬件线性化"方法，能直接在硬件中计算电路梯度，并通过器件级电路仿真实现了硬件端训练。该结果还可扩展到电阻-二极管网络级联以构建前馈等非对称网络，并引入由非理想二极管模型导出的新型 diode ReLU 激活函数。
正文
Abstract:It is shown that the port behavior of a resistor-diode network corresponds to the solution of a ReLU monotone operator equilibrium network (a neural network in the limit of infinite depth), giving a parsimonious construction of a neural network in analog hardware. We furthermore show that the gradient of such a circuit can be computed directly in hardware, using a procedure we call hardware linearization. This allows the network to be trained in hardware, which we demonstrate with a device-level circuit simulation. We extend the results to cascades of resistor-diode networks, which can be used to implement feedforward and other asymmetric networks. We finally show that different nonlinear elements give rise to different activation functions, and introduce the novel diode ReLU which is induced by a non-ideal diode model.
| Subjects: | Systems and Control (eess.SY); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE); Optimization and Control (math.OC) |
| MSC classes: | 65K10, 68T05, 93B30, 93D99 |
| Cite as: | arXiv:2509.13793 [eess.SY] |
| (or arXiv:2509.13793v3 [eess.SY] for this version) | |
| https://doi.org/10.48550/arXiv.2509.13793 arXiv-issued DOI via DataCite |
Submission history
From: Thomas Chaffey [view email]
[v1]
Wed, 17 Sep 2025 08:03:15 UTC (537 KB)
[v2]
Sat, 20 Jun 2026 08:54:53 UTC (567 KB)
[v3]
Wed, 7 Oct 2026 03:02:08 UTC (1,718 KB)
来源：arXiv：cs.LG · arxiv.org