We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Comparison of Models for Training Optical Matrix Multipliers in Neuromorphic PICs

Abstract: We experimentally compare simple physics-based vs. data-driven neural-network-based models for offline training of programmable photonic chips using Mach-Zehnder interferometer meshes. The neural-network model outperforms physics-based models for a chip with thermal crosstalk, yielding increased testing accuracy.
Comments: 3 pages, 3 figures
Subjects: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
Cite as: arXiv:2111.14787 [cs.LG]
  (or arXiv:2111.14787v1 [cs.LG] for this version)

Submission history

From: Ali Cem [view email]
[v1] Tue, 23 Nov 2021 12:15:21 GMT (3149kb,D)

Link back to: arXiv, form interface, contact.