Reproduce “Photonics for artificial intelligence and neuromorphic computing” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Bhavin J. Shastri, Alexander N. Tait, Thomas Ferreira de Lima, Wolfram H. P. Pernice, Harish Bhaskaran, C. David Wright, Paul R. Prucnal
The authors review photonic neural-network hardware that uses light for many-to-many connections and weighted sums, aiming to reduce communication bottlenecks in fast, bandwidth-intensive tasks. They distinguish optical-electrical-optical neurons, which regenerate output light, from all-optical neurons, whose outputs may be too weak to drive downstream units. A cited 2-fF device supports analog conversion, not a demonstrated system-level advantage over electronics.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.