An on-chip deep photonic neural network for image classification

  • Serre, T., Wolf, L., Bileschi, S., Riesenhuber, M. & Poggio, T. Robust object recognition with cortex-like mechanisms. IEEE Trans. Anal model. Mach. Information. 29411–426 (2007).

    Article

    Google Scholar

  • Wang, D., Su, J. & Yu, H. Feature extraction and natural language processing analysis for deep English language learning. IEEE Access 846335–46345 (2020).

    Article

    Google Scholar

  • Ribeiro, AH et al. Automatic diagnosis of 12-lead ECG using deep neural network. Nat. Common. 111760 (2020).

    ADS
    CASE
    Article

    Google Scholar

  • Lai, L. et al. Computer-aided diagnosis of pectus excavatum using CT images and deep learning methods. Science. representing ten20294 (2020).

    CASE
    Article

    Google Scholar

  • Yuan, B. et al. Unsupervised and supervised learning with neural network for human transcriptome analysis and cancer diagnosis. Science. representing ten19106 (2020).

    ADS
    CASE
    Article

    Google Scholar

  • Shin, H. et al. Deep convolutional neural networks for computer-aided sensing: CNN architectures, dataset characteristics and transfer learning. IEEE Trans. Med. Imaging 351285-1298 (2016).

    Article

    Google Scholar

  • Tajbakhsh, N. et al. Convolutional neural networks for the analysis of medical images: complete training or development? IEEE Trans. Med. Imaging 351299-1312 (2016).

    Article

    Google Scholar

  • LeCun, Y. & Bengio, Y. en The Brain and Neural Network Theory Handbook (ed. Arbib, MA) 255–258 (MIT Press, 1998).

  • LeCun, Y., Bengio, Y. & Hinton, G. Deep Learning. Nature 521436–444 (2015).

    ADS
    CASE
    Article

    Google Scholar

  • Barbastathis, G., Ozcan, A. & Situ, G. On the use of deep learning for computational imaging. Optical 6921–943 (2019).

    ADS
    Article

    Google Scholar

  • Krizhevsky, A., Sutskever, I. & Hinton, GE Imagenet Classification with Deep Convolutional Neural Networks. Adv. Neural information. Treat. System 251097-1105 (2012).

    Google Scholar

  • Nair, V. & Hinton, GE Ground Linear Units Improve Restricted Boltzmann Machines. In proc. 27th International Conference on Machine Learning (eds Fürnkranz, J. & Joachims, T.) 807–814 (Omnipress, 2010).

  • Ren, S., He, K., Girshick, R. & Sun, J. Faster R-CNN: Toward real-time object detection with region proposition networks. IEEE Trans. Anal model. Mach. Information. 391137-1149 (2017).

    Article

    Google Scholar

  • Li, H., Lin, Z., Shen, X., Brandt, J. & Hua, G. A cascade of convolutional neural networks for face detection. In 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)) 5325–5334 (IEEE, 2015).

  • Shen, Y. et al. Deep learning with coherent nanophotonic circuits. Nat. Photonics 11441–446 (2017).

    ADS
    CASE
    Article

    Google Scholar

  • Shastri, BJ et al. Photonics for artificial intelligence and neuromorphic computing. Nat. Photonics 15102–114 (2021).

    ADS
    CASE
    Article

    Google Scholar

  • Bogaerts, W. et al. Programmable photonic circuits. Nature 586207-216 (2020).

    ADS
    CASE
    Article

    Google Scholar

  • Moons, B. & Verhelst, M. A power-efficient, precision-scalable ConvNet processor in 40 nm CMOS. IEEE J. Solid State Circuits 52903–914 (2017).

    ADS
    Article

    Google Scholar

  • Lee, J. et al. UNPU: An energy-efficient deep neural network accelerator with fully variable bit precision. IEEE J. Solid State Circuits 54173–185 (2019).

    ADS
    Article

    Google Scholar

  • Hill, P et al. DeftNN: Dealing with bottlenecks for DNN execution on GPUs via synapse vector elimination and atrial compute data fission. In 2017 50th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO) 786–799 (IEEE, 2017).

  • Nurvitadhi, E. et al. Acceleration of binarized neural networks: comparison of FPGAs, CPUs, GPUs and ASICs. In 2016 International Conference on Field Programmable Technology (FPT) 77–84 (IEEE, 2016).

  • Ashtiani, F., Risi, A. & Aflatouni, F. Single-chip near-field nanophotonic imager. Optical 61255-1260 (2019).

    ADS
    CASE
    Article

    Google Scholar

  • Cheng, Z., Rios, C., Perince, WHP, Wright, CD & Bhaskaran, H. On-chip photonic synapse. Science. Adv. 3e1700160 (2017).

    ADS
    Article

    Google Scholar

  • Tait, AN et al. Neuromorphic photonic networks using silicon photonic weight banks. Science. representing seven7430 (2017).

    ADS
    Article

    Google Scholar

  • Feldmann, J. et al. All-optical spiked neurosynaptic networks with self-learning capabilities. Nature 569208-214 (2019).

    ADS
    CASE
    Article

    Google Scholar

  • Miscuglio, M. et al. All-optical nonlinear activation function for photonic neural networks. Opt. Mater. Express 83851–3863 (2018).

    ADS
    CASE
    Article

    Google Scholar

  • Jha, A., Huang, C. & Prucnal, PR Fully optical reconfigurable nonlinear activation functions for neuromorphic photonics. Opt. Lett. 454819–4822 (2020).

    ADS
    Article

    Google Scholar

  • Feldmann, J. et al. Parallel convolutional processing using an integrated photonic tensor core. Nature 58952–58 (2021).

    ADS
    CASE
    Article

    Google Scholar

  • Zuo, Y. et al. All-optical neural network with nonlinear activation functions. Optical 61132-1137 (2019).

    ADS
    CASE
    Article

    Google Scholar

  • Lin, X. et al. All-optical machine learning using diffractive deep neural networks. Science 3611004-1008 (2018).

    ADS
    MathSciNet
    CASE
    Article

    Google Scholar

  • Bueno, J. et al. Reinforcement learning in a large-scale photonic recurrent neural network. Optical 5756–760 (2018).

    ADS
    Article

    Google Scholar

  • Zhou, T. et al. Large-scale neuromorphic optoelectronic computing with a reconfigurable diffractive processing unit. Nat. Photonics 15367–373 (2021).

    ADS
    CASE
    Article

    Google Scholar

  • Chang, J. et al. Hybrid optical-electronic convolutional neural networks with optimized diffractive optics for image classification. Science. representing 812324 (2018).

    ADS
    Article

    Google Scholar

  • Xu, X et al. 11 TOPS photonic convolutional accelerator for optical neural networks. Nature 58944–51 (2021).

    ADS
    CASE
    Article

    Google Scholar

  • AMD RadeonMT RX 6700 XT graphics. https://www.amd.com/en/products/graphics/amd-radeon-rx-6700-xt.

  • Chollet, F. et al. Keras. https://keras.io (2015).

  • Tait, AN et al. Silicon photonic modulator neuron. Phys. Rev. Appl. 11064043 (2019).

    ADS
    CASE
    Article

    Google Scholar

  • Stone, M. Choice of cross-validation and evaluation of statistical predictions. JR statistics. Soc. Series B statistics. Methodology. 36111-147 (1974).

    MathSciNet
    MATH

    Google Scholar

  • Lecun, Y. et al. The MNIST dataset of handwritten digits. http://yann.lecun.com/exdb/mnist/ (1999).

  • Lecun, Y., Bottou, L., Bengio, Y. & Haffner, P. Gradient-based learning applied to document recognition. proc. IEEE 862278–2324 (1998).

    Article

    Google Scholar

  • Rakowski, M. et al. 45nm CMOS — Silicon Photonics Monolithic Technology (45CLO) for next-generation low-power, high-speed optical interconnects. In 2020 Optical Fiber Communications Conference and Exhibition (OFC) (IEEE, 2020).

  • Fahrenkopf, NM et al. AIM photonics MPW: a highly accessible state-of-the-art technology for rapid prototyping of photonic integrated circuits. IEEE J. Salt. High. Quantum electron. 251–6 (2019).

    Article

    Google Scholar

  • Borji, A., Cheng, M., Jiang, H. & Li, J. Protruding object detection: a reference. IEEE Trans. Image process. 245706–5722 (2015).

    ADS
    MathSciNet
    Article

    Google Scholar

  • Cheng, M., Mitra, NJ, Huang, X., Torr, PHS & Hu, S. Salient region detection based on global contrast. IEEE Trans. Anal model. Mach. Information. 37569-582 (2015).

    Article

    Google Scholar

  • Kist, AM Deep learning on advanced TPUs. Preprint at https://arxiv.org/abs/2108.13732 (2021).

  • Edge AI camera from IMAGO Technologies. https://imago-technologies.com/wp-content/uploads/2021/01/Specification-VisionAI-V1.2.pdf.

  • JeVois intelligent artificial vision. https://www.jevoisinc.com/collections/jevois-hardware/products/jevois-pro-deep-learning-smart-camera.

  • Kulyukin, V. et al. On the classification of images in the video analysis of the omnidirectional Apis mellifera traffic: random reinforced forests vs shallow convolutional networks. Appl. Science. 118141 (2021).

    CASE
    Article

    Google Scholar

  • Chiu, TY, Wang, Y. & Wang, H. A 3.7-43.7 GHz low-power variable-gain distributed amplifier in 90 nm CMOS. IEEE Microw. wirel. Making up. Lett. 31169-172 (2021).

    Article

    Google Scholar

  • Xuan, Z et al. A low-power 40 Gb/s silicon optical receiver. In 2015 IEEE Symposium on Radio Frequency Integrated Circuits (RFICs) 315–318 (IEEE, 2015).

  • Comments are closed.