Authors: George Sarantoglou, Adonis Bogris, Charis Mesaritakis

Published on: February 06, 2024

Impact Score: 8.22

Arxiv code: Arxiv:2402.03778

Summary

• What is new: A novel non-linear activation function for optical neural networks using a passive optical resonator, improving performance significantly.
• Why this is important: The need for more efficient non-linear activation functions in optical neural networks to enhance their computational power.
• What the research proposes: An integrated photonic non-linear activation function using a power independent, non-linear phase to amplitude conversion in a passive optical resonator.
• Results: Enhanced performance of time delayed reservoir computing (TDRC) in predicting the Santa Fe series by an order of magnitude compared to conventional methods.

Technical Details

Technological frameworks used: nan

Models used: Micro-Ring Resonators (MRRs), Time Delayed Reservoir Computing (TDRC)

Data used: Santa Fe series for one-step ahead prediction

Potential Impact

Optical computing, Data analytics firms, Companies in the field of AI and machine learning

Want to implement this idea in a business?

We have generated a startup concept here: PhotonBrain.