Nicolai Palm
Nicolai Palm

PhD Candidate in Computational Statistics & Data Science

About Me

Hi I’m Nico Palm. I am a mathematician and currently pursuing my PhD at the Munich Center for Machine learning (LMU) under the supervision of Thomas Nagler. Before that I worked for a couple of startups in the broad field of data analysis and machine learning. I specialize in the intersection of machine learning and statistics/mathematics: I develop statistical/mathematical methods that meet the requirements of modern data science, explore and prove their theoretical properties and package them in user-friendly software.

Interests
  • Theoretical Statistics
  • Uncertainty Quantification in ML
  • Category Theory Approaches in ML
  • Multi-Objective Optimization
Education
  • PhD Computational Statistics & Data Science

    LMU Munich

  • MSc Mathematics

    University of Regensburg

  • BSc Mathematics

    University of Regensburg

Research Interests

My current focus is on non-stationary time series. These are time series that can show trends and shifts in distribution. Non-stationary time series are difficult to study. As a result, very few general results are known. From a practical point of view, however, data are rarely stationary, which stresses the need for such results. My overall goal is to a) generalize fundamental theorems of probability theory (e.g. central limit theorems) and statistics and b) extend practical algorithms, such as the bootstrap, to the non-stationary context.

I am also interested in:

  • The broad area of uncertainty quantification for ML models, mainly from a frequentist perspective
  • Category theory in ML
  • Multi-objective optimization especially in the context of expensive blackbox functions (e.g. simulation or digital twin models)
  • Applying all this to real scenarios

Feel free to reach out!

Recent Publications