Peter Grünwald

Full Name
Prof.dr. P.D. Grünwald
Professor - Universiteit Leiden, Group leader, Scientific Staff Member
+31 20 592 4115
Machine Learning


Peter Grünwald’s research interests lie where statistics, computer science and information theory meet: theories of learning from data. He regularly publishes in the world’s top machine learning venues (NIPS, UAI, COLT,...) and statistics journals (Annals of Statistics, Journal of the Royal Statistical Society, ...). He has been co-program chair of both UAI (2010) and COLT (2015), and has been the general chair of UAI (2011). He is the author of the book The Minimum Description Length Principle, (MIT Press, 2007), which has become the standard reference for the MDL approach to learning from data. April 2010 he was co-awarded the Van Dantzig prize, the highest Dutch award in statistics and operations research. He has obtained NWO VIDI and VICI grants and, more recently (2016), an NWO TOP-1 grant.


My current research interests mostly focus on what I call Safe Learning, Safe Statistics and Safe Probability. The basic idea is to make sure that inference from data is done in - indeed - a safer way. The 'replicability crisis' in the applied sciences provides ample evidence that we very often jump to conclusions which simply aren't justified. The goal of much of my research is to improve this situation! Currently, I am mostly working on:

  • Safe Bayesian Inference: Reparing Bayesian inference under misspecification (when the model is wrong, but useful)
  • Safe Testing: Hypothesis Testing and Model Choice under Optional Stopping and Optional Continuation
  • Safe Probability: working with probability distributions that only capture parts, not all of your domain of interest.
  • Learning Bounds: quantifying how many data are needed to reach conclusions of a desired quality in machine learning and sequential prediction, with generalized Bayesian methods, PAC-Bayesian methods and MDL methods.


Current projects with external funding

  • Enabling Personalized Interventions (EPI)
  • Safe Bayesian Inference: A Theory of Misspecification based on Statistical Learning (SAFEBAYES)

Professional activities

  • Member: International Society for Bayesian Analysis (ISBA)
  • Professor: Universiteit van Leiden
  • Editor: Statistica Neerlandica
  • President: Association for Computational Learning


  • NWO TOP Module 1 grant (2 PhD students, 1 postdocs) ᅠ (2016)
  • Vici Innovational Research Grant NWO (2010)
  • Vidi Innovational Research Grant NWO (2005)


  • Van Dantzig Award (2010)