Deep learning has been transforming our ability to execute advanced inference tasks using computers. Here we introduce a physical mechanism to perform machine learning by demonstrating an all-optical diffractive deep neural network (DNN) architecture that can implement various functions following the deep learning-based design of passive diffractive layers that work collectively. We created 3D-printed DNNs that implement classification of images of handwritten digits and fashion products, as well as the function of an imaging lens at a terahertz spectrum. Our all-optical deep learning framework can perform, at the speed of light, various complex functions that computer-based neural networks can execute; will find applications in all-optical image analysis, feature detection, and object classification; and will also enable new camera designs and optical components that perform distinctive tasks using DNNs.

Even though the terahertz spectrum is well suited for chemical identification, material characterization, biological sensing and medical imaging, practical development of these applications has been hindered by attributes of existing terahertz optoelectronics. Here we demonstrate that the use of plasmonic contact electrodes can significantly mitigate the lowquantum efficiency performance of photoconductive terahertz optoelectronics. The use of plasmonic contact electrodes offers nanoscale carrier transport path lengths for the majority of photocarriers, increasing the number of collected photocarriers in a subpicosecond timescale and, thus, enhancing the optical-to-terahertz conversion efficiency of photoconductive terahertz emitters and the detection sensitivity of photoconductive terahertz detectors. We experimentally demonstrate 50 times higher terahertz radiation powers from a plasmonic photoconductive emitter in comparison with a similar photoconductive emitter with non-plasmonic contact electrodes, as well as 30 times higher terahertz detection sensitivities from a plasmonic photoconductive detector in comparison with a similar photoconductive detector with non-plasmonic contact electrodes.

Deep learning has been transformative in many fields, also motivating the emergence of various optical computing architectures. Diffractive optical network is a recently-introduced optical computing framework that merges wave optics with deep learning methods to design optical neural networks. Diffraction-based all-optical object recognition systems, designed through this framework and fabricated by 3D printing, have been reported to recognize hand-written digits and fashion products, demonstrating all-optical inference and generalization to sub-classes of data. These previous diffractive approaches employed monochromatic coherent light as the illumination source. Here, we report a broadband diffractive optical neural network design that simultaneously processes a continuum of wavelengths generated by a temporally-incoherent broadband source to all-optically perform a specific task learned using deep learning. We experimentally validated the success of this broadband diffractive neural network architecture by designing, fabricating and testing seven different multi-layer, diffractive optical systems that transform the optical wavefront generated by a broadband THz pulse to realize (1) a series of tunable, single passband as well as dual passband spectral filters, and (2) spatially-controlled wavelength de-multiplexing. Merging the native or engineered dispersion of various material systems with a deep learning-based design strategy, broadband diffractive neural networks help us engineer light-matter interaction in 3D, diverging from intuitive and analytical design methods to create task-specific optical components that can all-optically perform deterministic tasks or statistical inference for optical machine learning.