Webinar 'Photonics & Artificial Intelligence'
Tuesday 9 February
15:00 - 17:00
Title: Photonic reservoir computing and
its implementation using delay-based systems
The concept of reservoir computing (RC), a paradigm within neuromorphic computing, offers a framework to exploit the transient dynamics within a neural network for performing useful computation. RC simplifies the training procedure for neural networks, by keeping the neural network fixed and relying on a trained output layer to generate the desired output signals. In this presentation, the basic operation of a (photonic) RC system will be discussed. We will then focus on delay-based photonic RC, which has gained considerable interest as it allows for simple technological implementations of the RC concept that can operate at high speed.
Guy Verschaffelt is currently professor at the Applied Physics research group of the VUB. His research is focused on polarization and noise properties of semiconductor lasers, speckle and speckle reduction techniques, the study of non-linear dynamics in photonic systems and photonic implementations of reservoir computing and Ising machines. He is currently secretary of the board of the IEEE Photonics Society Benelux Chapter. He is co-author of more than 150 journal and conference papers, and teaches to bachelor and master students in Science and Engineering.
Research group description:
The research of the Applied Physics research group (APHY) of the Vrije Universiteit Brussel is focused on understanding the physical principles of operation and the dynamical properties of electromagnetic, opto-electronic and photonic devices. APHY is an interdisciplinary group that borrows advanced methods from physics – from nonlinear dynamics, stochastic processes, complex systems, electromagnetism, general relativity etc. – and applies them to diverse problems. A lot of attention is given to the study of the dynamics of micro-semiconductor lasers and of opto-electronic systems with delayed optical feedback, and the use of these systems for reservoir computing, machine learning and the implementation of Ising machines.