arXiv：cs.LG· Sowjanya Tammali, Wilkie Olin-Ammentorp·· 4 小时前AI 评分30
Lock-in EP：面向振荡模拟硬件的原位训练算法
Lock-in EP: An In-Situ Training Algorithm for Oscillatory Hardware
AI 导读
研究者提出并验证了 lock-in equilibrium propagation（LIEP）训练方法，可为振荡网络中每个元件提供局部梯度信息，无需独立的前向与反向扫描，从而有望在模拟振荡硬件平台上实现在线学习。LIEP 既可用于从零训练，也能在预训练参数被扰动后恢复性能，并可表述为三因子更新规则。该方法目前仅在浅层网络上得到验证，作者认为其他架构或可将其扩展到深度、大规模网络。
正文
Abstract:Analog hardware platforms offer the potential to reduce energy consumption over digital architectures, but in order to succeed, large-scale analog systems must also be able to operate with or recover from the variability of their components. Towards this goal, we derive and demonstrate the lock-in equilibrium propagation (LIEP) training method. LIEP provides local gradient information for each component in an oscillatory network without separate forward and backward sweeps, potentially allowing for in-situ learning capabilities on analog oscillatory hardware platforms. We demonstrate that LIEP can be used both for ab-initio training as well as recovering performance when pre-trained parameters are perturbed. We show that LIEP can be formulated as a three-factor update rule, and suggest that although the method is currently only validated on shallow networks, alternate architectures may allow it to extend to deep and large-scale networks addressing complex tasks.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.07283 [cs.LG] |
| (or arXiv:2610.07283v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07283 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Wilkie Olin-Ammentorp [view email]
[v1]
Mon, 5 Oct 2026 19:20:18 UTC (234 KB)
来源：arXiv：cs.LG · arxiv.org