Niclas Dern

Mathematics undergraduate at the Technical University of Munich

prof_pic.jpeg

I am a 3rd-year mathematics undergraduate student at the Technical University of Munich, currently visiting UC Berkeley supported by the German Academic Scholarship Foundation. My research motivation is to make large-scale machine learning systems more robust, interpretable, and trustworthy, ideally in a principled manner.

In the past, I’ve interned at the Vector Institute, where I worked with Geoff Pleiss on theoretical properties of ensembles, and worked at the Helmholtz AI Institute with Niki Kilbertus and Elisabeth Ailer on estimating nonlinear causal effects in confounded environments. As I have a passion for EdTech, I also co-founded Mathis, an AI tutor that provides individualized feedback for math exams to German high school students.

I am an Atlas Fellow (2022) and a winner of the National Student Championships in Computer Science (Germany, 2020). In my free time, I enjoy reading, swimming, traveling, and meeting new people.

news

Dec 17, 2024 Our paper Effects of Adaptive Feedback Generated by a Large Language Model: A Case Study in Teacher Education was published in Computers and Education: Artificial Intelligence.
Oct 21, 2024 New preprint online: Theoretical Limitations of Ensembles in the Age of Overparameterization. We show that ensembles of overparameterized models behave similarly to single, larger models.
Sep 25, 2024 Our paper Targeted Sequential Indirect Experiment Design is accepted to NeurIPS 2024.
Sep 19, 2024 Our submission won first prize in the competition Learning with AI: Intelligent Education for an Intelligent Future organized by the German Initiative for Network Information.

selected publications

  1. limitations-ensembles.png
    Theoretical Limitations of Ensembles in the Age of Overparameterization
    Niclas Dern, John P Cunningham, and Geoff Pleiss
    arXiv preprint arXiv:2410.16201, 2024
  2. targeted-experiments.png
    Targeted Sequential Indirect Experiment Design
    Elisabeth Ailer, Niclas Dern, Jason Hartford, and 1 more author
    NeurIPS 2024, 2024
  3. sumformer.png
    Sumformer: Universal approximation for efficient transformers
    Silas Alberti, Niclas Dern, Laura Thesing, and 1 more author
    In Topological, Algebraic and Geometric Learning Workshops 2023, 2023