Yibo Dong, Dajun Lin, Long Chen, Baoli Li, Xi Chen, Qiming Zhang, Haitao Luan, Xinyuan Fang, Min Gu
{"title":"Compact eternal diffractive neural network chip for extreme environments","authors":"Yibo Dong, Dajun Lin, Long Chen, Baoli Li, Xi Chen, Qiming Zhang, Haitao Luan, Xinyuan Fang, Min Gu","doi":"10.1038/s44172-024-00211-6","DOIUrl":null,"url":null,"abstract":"Artificial intelligence applications in extreme environments place high demands on hardware robustness, power consumption, and speed. Recently, diffractive neural networks have demonstrated superb advantages in high-throughput light-speed reasoning. However, the robustness and lifetime of existing diffractive neural networks cannot be guaranteed, severely limiting their compactness and long-term inference accuracy. Here, we have developed a millimeter-scale and robust bilayer-integrated diffractive neural network chip with virtually unlimited lifetime for optical inference. The two diffractive layers with binary phase modulation were engraved on both sides of a quartz wafer. Optical inference of handwritten digital recognition was demonstrated. The results showed that the chip achieved 82% recognition accuracy for ten types of digits. Moreover, the chip demonstrated high-performance stability at high temperatures. The room-temperature lifetime was estimated to be 1.84×1023 trillion years. Our chip satisfies the requirements for diffractive neural network hardware with high robustness, making it suitable for use in extreme environments. Yibo Dong et al. implement a compact and robust diffractive neural network chip with a virtually unlimited lifetime for optical inference. The chip demonstrates high accuracy and high stability even after high temperature aging, aiming at applications in extreme environments.","PeriodicalId":72644,"journal":{"name":"Communications engineering","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s44172-024-00211-6.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Communications engineering","FirstCategoryId":"1085","ListUrlMain":"https://www.nature.com/articles/s44172-024-00211-6","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Artificial intelligence applications in extreme environments place high demands on hardware robustness, power consumption, and speed. Recently, diffractive neural networks have demonstrated superb advantages in high-throughput light-speed reasoning. However, the robustness and lifetime of existing diffractive neural networks cannot be guaranteed, severely limiting their compactness and long-term inference accuracy. Here, we have developed a millimeter-scale and robust bilayer-integrated diffractive neural network chip with virtually unlimited lifetime for optical inference. The two diffractive layers with binary phase modulation were engraved on both sides of a quartz wafer. Optical inference of handwritten digital recognition was demonstrated. The results showed that the chip achieved 82% recognition accuracy for ten types of digits. Moreover, the chip demonstrated high-performance stability at high temperatures. The room-temperature lifetime was estimated to be 1.84×1023 trillion years. Our chip satisfies the requirements for diffractive neural network hardware with high robustness, making it suitable for use in extreme environments. Yibo Dong et al. implement a compact and robust diffractive neural network chip with a virtually unlimited lifetime for optical inference. The chip demonstrates high accuracy and high stability even after high temperature aging, aiming at applications in extreme environments.