Zheng Hanyu, Liu Quan, Zhou You, Kravchenko Ivan I, Huo Yuankai, Valentine Jason
Department of Electrical and Computer Engineering, Vanderbilt University, Nashville, TN 37212, USA.
Department of Computer Science, Vanderbilt University, Nashville, TN 37212, USA.
Sci Adv. 2022 Jul 29;8(30):eabo6410. doi: 10.1126/sciadv.abo6410. Epub 2022 Jul 27.
Rapid advances in deep learning have led to paradigm shifts in a number of fields, from medical image analysis to autonomous systems. These advances, however, have resulted in digital neural networks with large computational requirements, resulting in high energy consumption and limitations in real-time decision-making when computation resources are limited. Here, we demonstrate a meta-optic-based neural network accelerator that can off-load computationally expensive convolution operations into high-speed and low-power optics. In this architecture, metasurfaces enable both spatial multiplexing and additional information channels, such as polarization, in object classification. End-to-end design is used to co-optimize the optical and digital systems, resulting in a robust classifier that achieves 93.1% accurate classification of handwriting digits and 93.8% accuracy in classifying both the digit and its polarization state. This approach could enable compact, high-speed, and low-power image and information processing systems for a wide range of applications in machine vision and artificial intelligence.
深度学习的快速发展已在从医学图像分析到自主系统等多个领域引发了范式转变。然而，这些进展导致数字神经网络具有巨大的计算需求，从而导致高能耗以及在计算资源有限时实时决策方面的局限性。在此，我们展示了一种基于超光学的神经网络加速器，它可以将计算成本高昂的卷积运算卸载到高速、低功耗的光学系统中。在这种架构中，超表面在目标分类中实现了空间复用以及诸如偏振等额外信息通道。采用端到端设计来共同优化光学和数字系统，从而得到一个强大的分类器，该分类器对手写数字的分类准确率达到93.1%，对数字及其偏振状态进行分类的准确率达到93.8%。这种方法可为机器视觉和人工智能中的广泛应用实现紧凑、高速且低功耗的图像和信息处理系统。
