High-performance real-world optical computing trained by in situ model-free optimization

arXiv:2307.11957v5 Announce Type: replace-cross
Abstract: Optical computing systems provide high-speed and low-energy data processing but face deficiencies in computationally demanding training and simulation-to-reality gaps. We propose a gradient-based model-free optimization (G-MFO) method based on a Monte Carlo gradient estimation algorithm for computationally efficient in situ training of optical computing systems. This approach treats an optical computing system as a black box and back-propagates the loss directly to the optical computing weights' probability distributions, circumventing the need for a computationally heavy …

abstract algorithm arxiv computing computing systems cs.cv cs.et cs.lg data data processing energy face free gradient low low-energy optical optical computing optimization performance physics.optics processing reality simulation speed systems training type world