arXiv · 2401.03564
Experimental Demonstration of Imperfection-Agnostic Local Learning Rules on Photonic Neural Networks with Mach-Zehnder Interferometric Meshes
Abstract
Mach-Zehnder Interferometric meshes are attractive for low-loss photonic matrix multiplication but are challenging to program. Using least-squares optimization of directional derivatives, we experimentally demonstrate that desired matrix updates can be implemented agnostic to hardware imperfections. \c{opyright} 2024 The Author(s)
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Luis El Srouji, Mehmet Berkay On, Yun-Jhu Lee, Mahmoud Abdelghany, S. J. Ben Yoo. 2024-01-07. Experimental Demonstration of Imperfection-Agnostic Local Learning Rules on Photonic Neural Networks with Mach-Zehnder Interferometric Meshes. https://arxiv.org/abs/2401.03564
Cite the original work for its findings. Save a collection to share your selection of sources.