要素不精确的光学神经网络的设计（Emerging Technologies）

• 2020 年 1 月 8 日
• 筆記

为了设计可伸缩的、抗故障的光学神经网络，我们研究了结构设计对光学神经网络在非精确元件方面的稳健性的影响。我们设计了两个光学神经网络来对手写数字进行分类，一个具有更好的可调设计(GridNet)，另一个具有更好的容错能力(FFTNet)。在没有任何缺陷的情况下进行模拟时，GridNet的精确度(∼98%)比FFTNet(∼95%)好。然而，在其光子分量误差较小的情况下，容错能力越强的FFTNet能超越GridNet。我们还对光学神经网络关于不同程度和不同类型的缺陷的敏感性进行了深入的定量和定性分析。本文的研究结果为网络容错的原则设计提供了指导，同时也为下一步的研究奠定了基础。

原文题目：Design of optical neural networks with component imprecisions

原文： For the benefit of designing scalable, fault resistant optical neural networks (ONNs), we investigate the effects architectural designs have on the ONNs’ robustness to imprecise components. We train two ONNs – one with a more tunable design (GridNet) and one with better fault tolerance (FFTNet) – to classify handwritten digits. When simulated without any imperfections, GridNet yields a better accuracy (～ 98%) than FFTNet (～ 95%). However, under a small amount of error in their photonic components, the more fault tolerant FFTNet overtakes GridNet. We further provide thorough quantitative and qualitative analyses of ONNs’ sensitivity to varying levels and types of imprecisions. Our results offer guidelines for the principled design of fault-tolerant ONNs as well as a foundation for further research.

.

原文作者：Michael Y.-S. Fang, Sasikanth Manipatruni, Casimir Wierzynski, Amir Khosrowshahi, Michael R. DeWeese

原文地址： https://arxiv.org/abs/2001.01681