Photonic metamaterials have fascinating properties for light, with many applications from communication and imaging to photovoltaics and topological physics. They are however notoriously difficult to design, owing to the large parameter space and multidimensional design targets. We want to investigate how artificial intelligence could predict the properties of metamaterials that include two-dimensional metasurfaces and photonic crystals, how it could help us understand the inner workings of these systems, and how it could be used to design metamaterial devices with targeted functionality. Specifically, we study the incorporation of rigorous self-consistent physical models into the machine learning method to achieve those goals.
The Photonic Forces group at AMOLF, led by prof. Ewold Verhagen, seeks a motivated and talented postdoc to undertake this project. The project lies at the interface of electromagnetic modelling and theory and deep learning methods. You will develop methods and test them towards the inverse design of state-of-art photonic systems, in collaboration with international academic and industrial partners. You will be involved in all aspects of development, design, and dissemination of the results.