Training with latent full-precision weights preserves model accuracy.
DoReFa quantizers fine-tune input and kernel precision for photonic use.
The result: compact, fast, and robust AI hardware.
A pragmatic route to efficient optical AI.
🌟 Follow us on LinkedIn! https://www.linkedin.com/company/clever-project/?viewAsMember=true
🌟 Check the updates from the website: www.cleverproject.eu 🌟 Full paper in: https://zenodo.org/records/11045819
We use cookies to ensure that we give you the best experience on our website.