Researchers at the George Washington University, together with researchers at the University of California, Los Angeles, and the deep-tech venture startup Optelligence LLC, have developed an optical convolutional neural network accelerator capable of processing large amounts of information, on the order of petabytes, per second.
Machine learning hardware requires a huge amount of computing power supplies. Graphics and tensor processing accelerators help address this, but are intrinsically challenged by serial data processing that requires iterative data processing and encounters delays from wiring and circuit constraints. Optical alternatives could also 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.
Researchers achieved a breakthrough in optical machine learning by replacing 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.
Unlike the current paradigm in electronic machine learning hardware that processes information sequentially, this processor uses the Fourier optics, a concept of frequency filtering which allows for performing the required convolutions of the neural network as much simpler element-wise multiplications using the digital mirror technology.