Affiliation:
1. Electrical and Computer Engineering Department University of California Los Angeles Los Angeles CA 90095 USA
2. Bioengineering Department University of California Los Angeles Los Angeles CA 90095 USA
3. California NanoSystems Institute (CNSI) University of California Los Angeles Los Angeles CA 90095 USA
Abstract
Diffractive deep neural networks (D2NNs), comprised of spatially engineered passive surfaces, collectively process optical input information at the speed of light propagation through a thin diffractive volume, without any external computing power. Diffractive networks were demonstrated to achieve all‐optical object classification and perform universal linear transformations. Herein, a “time‐lapse” image classification scheme using a diffractive network is demonstrated for the first time, significantly advancing its classification accuracy and generalization performance on complex input objects by using the lateral movements of the input objects and/or the diffractive network, relative to each other. In a different context, such relative movements of the objects and/or the camera are routinely being used for image super‐resolution applications; inspired by their success, a time‐lapse diffractive network is designed to benefit from the complementary information content created by controlled or random lateral shifts. The design space and performance limits of time‐lapse diffractive networks are numerically explored, revealing a blind testing accuracy of 62.03% on the optical classification of objects from the CIFAR‐10 dataset. This constitutes the highest inference accuracy achieved so far using a single diffractive network on the CIFAR‐10 dataset. Time‐lapse diffractive networks will be broadly useful for the spatiotemporal analysis of input signals using all‐optical processors.
Funder
U.S. Department of Energy
Reference42 articles.
1. Machine learning: Trends, perspectives, and prospects
2. Deep learning
3. W.Knight https://www.wired.com/story/ai-great-things-burn-planet/(accessed: August 2022).
4. R.Toews https://www.forbes.com/sites/robtoews/2020/06/17/deep-learnings-climate-change-problem/(accessed: August 2022).
5. The Future of Deep Learning Is Photonic: Reducing the energy needs of neural networks might require computing with light
Cited by
11 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献