Photonics and AI - 3Virtual room: Photons Canada - 4
|Thursday, May 28|
AI-3-28-1 / Quantum photonic processors to accelerate machine learning
* Jacques Carolan, University of Copenhagen, Denmark
The generation and manipulation of quantum states of light has historically played a critical role in the development of quantum information science: from the first violation of Bells inequality to the more recent development of near-term quantum algorithms such as the variational quantum eigensolver. In this talk, I present a new frontier for photons at the intersection of quantum mechanics and machine learning. I will first provide a short introduction to the field of quantum photonics, then demonstrate how quantum photonic processors can accelerate both quantum and classical machine learning. Finally, I show how optimization techniques can enhance large-scale quantum control and provide a new path towards efficient verification of near-term quantum processors.
AI-3-28-2 / Silicon Photonic Neural Networks and Applications
* Bhavin Shastri, Queen's University, Canada
Bicky Marquez, Queen's University
Alexander Tait, Princeton University
Thomas Ferreira de Lima, Princeton University
Hsuan-Tung Peng, Princeton University
Chaoran Huang, Princeton University
Paul Prucnal, Princeton University
Neuromorphic photonic processors promise orders of magnitude improvements in both speed and energy efficiency over purely digital electronic approaches. We will provide an overview of neuromorphic photonic systems and their application to machine learning and specifically deep learning inference with a hybrid digital electronics and analog photonics architecture based on silicon photonics. We will discuss scalability in the context of designing a full-scale neuromorphic photonic processing system, considering aspects such as signal integrity, noise, and hardware fabrication platforms.