Lars Mikelsons
Diploma in Mathematics. PhD in Mechatronics. Worked in Bosch Corporate Research now head of the Chair for Mechatronics at the University of Augsburg.
Sessions
If there is something YOU know about a physical system, AI shouldn’t need to learn it. How to integrate your system knowledge into an ML development process is the core topic of this hands-on workshop. The entire workshop evolves around a challenging use case from robotics: Modeling a robot that is able to write arbitrary messages with a pen. After introducing the topic and the considered use case, participants can experiment with their very own robot model.
We introduce MeshGraphNets.jl, an open-source library that provides the neural network architecture of MeshGraphNets by Google DeepMind in the context of NeuralODEs. It supports full computation on the GPU for accelerated training and evaluation of the system and provide an interface for customizing MeshGraphNets for your own use case. We show that the network has an improved understanding of the underlying phenomena on a network derived from a real world hydraulic brake system.