Optical Sciences Centre, Swinburne University of Technology, Hawthorn, Victoria, Australia.
Electro-Photonics Laboratory, Department of Electrical and Computer Systems Engineering, Monash University, Clayton, Victoria, Australia.
Nature. 2021 Jan;589(7840):44-51. doi: 10.1038/s41586-020-03063-0. Epub 2021 Jan 6.
Convolutional neural networks, inspired by biological visual cortex systems, are a powerful category of artificial neural networks that can extract the hierarchical features of raw data to provide greatly reduced parametric complexity and to enhance the accuracy of prediction. They are of great interest for machine learning tasks such as computer vision, speech recognition, playing board games and medical diagnosis. Optical neural networks offer the promise of dramatically accelerating computing speed using the broad optical bandwidths available. Here we demonstrate a universal optical vector convolutional accelerator operating at more than ten TOPS (trillions (10) of operations per second, or tera-ops per second), generating convolutions of images with 250,000 pixels-sufficiently large for facial image recognition. We use the same hardware to sequentially form an optical convolutional neural network with ten output neurons, achieving successful recognition of handwritten digit images at 88 per cent accuracy. Our results are based on simultaneously interleaving temporal, wavelength and spatial dimensions enabled by an integrated microcomb source. This approach is scalable and trainable to much more complex networks for demanding applications such as autonomous vehicles and real-time video recognition.
卷积神经网络受生物视觉皮层系统启发，是一类强大的人工神经网络，可以提取原始数据的分层特征，提供大大降低的参数复杂性，并提高预测的准确性。它们在机器学习任务中非常有趣，例如计算机视觉、语音识别、玩棋盘游戏和医学诊断。光神经网络有望利用可用的广泛光带宽极大地提高计算速度。在这里，我们展示了一种通用的光学向量卷积加速器，其运行速度超过每秒 10 万亿次操作（每秒 10 太次操作，或 tera-ops 每秒），可对 25 万个像素的图像进行卷积——足以用于面部图像识别。我们使用相同的硬件顺序形成具有 10 个输出神经元的光学卷积神经网络，以 88%的准确率成功识别手写数字图像。我们的结果基于集成微梳源支持的同时交错时间、波长和空间维度。这种方法是可扩展的，可以针对更复杂的网络进行训练，以满足自动驾驶汽车和实时视频识别等要求苛刻的应用。
