In the post-Moore era, AI × Photonics is a novel frontier that establishes the area of optical data science. By leveraging diverse AI computing architectures together with cutting-edge photonics technology, this talk introduces the photonic accelerator (PAXEL) and its use cases.

The talk will focus on “Small-world Network (SWN)” based reservoir computing (RC) as the computing engine of PAXEL. RC has gained attention due to its drastically simplified training. SWN is characterized by highly-clustered nodes with a short path length, making it highly effective for tasks such as optical signal equalization, fiber sensing, laser chaotic signal analysis, and human vital sign detection.

A systematic synthesis approach will be introduced for optimizing the reservoir architecture using SWN's controllable parameters. The talk will conclude with a concept of minimal design using multi-stage SWN reservoirs for photonic chiplets, paving the way for future PAXEL photonic integrated circuits (PICs).

About the Speaker

Prof. Ken-ichi Kitayama is Professor Emeritus of Osaka University and currently serves as R&D Advisor at the Network Research Institute, National Institute of Information and Communications Technology (NICT), Tokyo, Japan. He is also a Research Fellow at the Central Research Laboratory, Hamamatsu Photonics.