Research Seminar on AI: Advanced Beam Shaping for Laser Materials Processing Based on Diffractive Neural Networks

 

Dienstag, 06.09.2022, 16.00 Uhr

Diffractive neural networks are a physical realization of artificial neural networks with light acting as the information and surfaces with a trained microstructure (so called diffractive optical elements) acting as the layers. These networks have been demonstrated to solve traditional image processing tasks at the speed of light, containing only passive (no energy consumption) optical components. In this talk, we will explain the physical principles, discuss our own implementation and demonstrate the capabilities of diffractive neural networks as a powerful laser beam shaping tool.

Paul Buske is a PhD student at the Chair for Technology of Optical Systems (TOS) supervised by Prof. Carlo Holly. His research focuses on wave optical simulations, laser beam shaping and smart imaging system.