CRC 1459 - C05: Coherent nanophotonic neural networks with adaptive molecular systems

Basic data for this project

Description

The goal of this project is to develop and evaluate fundamental building blocks for optical artificial neural networks based on coherent nanophotonic circuits. The implementation of effective nonlinear photo-responsive systems has emerged as a primary challenge for novel realizations of optical neural networks for deep-learning applications. By comparing different systems, considering both optical-computing and computer-science aspects, a variety of requirements for nanophotonic neural networks will be determined. Crucial challenges in achieving light-matter interaction at the microscale that is strong enough to yield the nonlinear optical behaviour required for activation functions in optical neural networks will be addressed.