Jonas is building a group for safety- & efficiency- aligned learning. Before this, he has spent time at the University of Maryland and the University of Siegen.
His main goal is to investigate the safety and efficiency of modern machine learning. There are a number of fundamental machine learning questions that come up in these topics that we still do not understand well.
In safety, these are questions such as the principles of data poisoning, the subtleties of watermarking for generative models, privacy questions in federated learning, or adversarial attacks against large language models. Can we ever make these models “safe”, and how do we define this? Are there feasible technical solutions that reduce harm?
Further, he is researching the efficiency of modern AI systems, especially for large language models. How efficient can we make these systems, can we train strong models with little compute? Can we extend the capabilities of language models with recursive computation? How do efficiency modifications impact the safety of these models?