Phys.org December 18, 2020
Optical alternatives to electronic hardware could help speed up machine learning processes by simplifying the way information is processed in a non-iterative way. However, photonic-based machine learning is typically limited by the number of components that can be placed on photonic integrated circuits, limiting the interconnectivity, while free-space spatial-light-modulators are restricted to slow programming speeds. A team of researchers in the US (George Washington University, UCLA, industry) replaced spatial light modulators with digital mirror-based technology, thus developing a system over 100 times faster. The non-iterative timing of this processor, in combination with rapid programmability and massive parallelization, enables this optical machine learning system to outperform even the top-of-the-line graphics processing units by over one order of magnitude, with room for further optimization beyond the initial prototype. The processor uses Fourier optics which allows for performing the required convolutions of the neural network as much simpler element-wise multiplications using the digital mirror technology. This prototype demonstration shows a commercial path for optical accelerators ready for several applications like network-edge processing, datacenters and high-performance computer systems…read more. Open Access TECHNICAL ARTICLE