Paper Reading Group - Accelerating Physical Simulations with Graph Neural Networks

We are happy to welcome the next round of our AI paper reading group, where we will cover AI-related papers to keep up to date with what is happening in this fast-moving environment 🚀

In this edition of the Paper Reading Group, we welcome Winfried Ripken, Machine Learning Researcher at Merantix Momentum, to discuss an interesting intersection of machine learning and physics.

We specifically look at how Graph Neural Networks can be trained on data generated by a Finite Element Simulation. We will first discuss the basic properties of GNNs that make them a suitable neural network architecture for such a task. In the following we will touch upon some recent papers and talk about challenges for state-of-the-art models in this exciting domain.

Register below to attend the event in person or join online here 💻