"Emerging topics in Photonic AI"

Abstract:

Neuromorphic photonic systems are emerging as an attractive alternative for realization of certain types of artificial intelligence systems, mainly due to their potential for low-energy consumption and extremely fast responses. In this talk, I will present an introduction into the operation principles of such photonic systems, starting with a basic introduction of artificial neural networks and then optical neural networks, with a special focus on diffractive optical neural networks. I will then give an overview of some of the existing concepts, research directions, and also challenges in the field of photonic AI, by introducing the various utilized photonic platforms, training approaches, underlying physical architectures, and approaches for realizing nonlinear activations. 

Bio:

Dr. Sina Saravi has completed his B.Sc. in Electrical Engineering in Iran, and then his M.Sc. in Photonics and doctoral studies in Physics at the Friedrich Schiller University of Jena, Germany, followed by a Postdoc period in Jena. He is currently leading a junior research group supported by the Carl Zeiss Foundation at the Institute of Applied Physics in Jena. He will soon start as a professor at the Department of Electrical Engineering and Information Technology of Paderborn University in Germany.