PeopleofSuperintelligence

Researcher

Peter Grünwald

Peter Grünwald is a machine learning researcher whose work connects data compression with statistical inference. His 1998 PhD dissertation at the University of Amsterdam was titled The Minimum Description Length Principle and Reasoning under Uncertainty, and his 2007 book introduced MDL for learning from data. In 2008, he became Professor of Statistical Learning at Leiden; he shared the Van Dantzig Prize in 2010. From 2016 to 2023, he led CWI's Machine Learning group and coauthored Safe Testing in 2019; an ERC Advanced Grant supported his e-value research in 2024, and he became an IMS Fellow in 2025. As of October 2026, he works in CWI's Machine Learning group, holds a part-time professorship at Leiden, serves on CWI's scientific management team, and is an ELLIS Fellow.

Born
1970
Works
287
Citations
4,488
h-index
32
Badges

Years link to their sources. Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 1998

    An MDL dissertation

    In 1998, his PhD dissertation at the University of Amsterdam is titled The Minimum Description Length Principle and Reasoning under Uncertainty.

  2. 2007

    The Minimum Description Length Principle

    In 2007, he publishes The Minimum Description Length Principle, a comprehensive introduction to MDL that uses data compression to select models across statistics and machine learning.

  3. 2008

    Professor of Statistical Learning

    In November 2008, he becomes Professor of Statistical Learning at the Mathematical Institute of Leiden University and takes up the post one day a week.

  4. 2010

    The Van Dantzig Prize

    He shares the 2010 Van Dantzig Prize, the highest Dutch award in statistics and operations research.

  5. 2016–2023

    CWI leadership and Safe Testing

    From 2016 to 2023, he heads CWI's Machine Learning group. In 2019, he coauthors Safe Testing with Rianne de Heide and Wouter Koolen; the paper develops e-value-based hypothesis tests that preserve Type-I error guarantees when researchers decide to run further studies based on previous outcomes.

  6. 2024

    An ERC grant for e-values

    In 2024, he receives a €2.5 million ERC Advanced Grant to develop a more flexible theory of statistics based on e-values.

  7. 2025

    An IMS Fellowship

    The Institute of Mathematical Statistics elects him a Fellow in 2025 for work on e-values, Bayesian inference under model misspecification, and minimum description length learning.

  8. As of October 2026

    CWI, Leiden and ELLIS roles

    As of October 2026, he works in CWI's Machine Learning group and holds a part-time professorship at Leiden University. CWI lists him as a scientific member of its management team, and ELLIS lists him as a Fellow.

Career

  1. Since 2016NWO-IHead of Machine Learning group
  2. Since 2008Leiden UniversityProfessor of Statistical Learning
  3. 1988–2008Universität Hamburg

Education

  1. 1998University of AmsterdamPhD

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

Most-cited works

  1. Game theory, maximum entropy, minimum discrepancy and robust Bayesian decision theory2004 · The Annals of Statistics · 454 citationsWith A. P. Dawid
  2. Advances in Minimum Description Length: Theory and Applications2005 · 403 citations
  3. A tutorial introduction to the minimum description length principle2004 · 251 citations
  4. The Minimum Description Length Principle (Adaptive Computation and Machine Learning)2007 · The MIT Press eBooks · 176 citations
  5. Model Selection Based on Minimum Description Length2000 · Journal of Mathematical Psychology · 163 citations

From OpenAlex, as of 21 October 2026.

  1. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes Leiden University. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.
  2. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Leiden University. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  3. Google has the highest combined h-index among AI organizationsIncludes Leiden University. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.
  4. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes Leiden University. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Peter Grünwald yet. Send us a free-licensed photo.

  1. Peter Grunwald's publications homepages.cwi.nl. Checked 20 October 2026.
  2. The Minimum Description Length Principle homepages.cwi.nl. Checked 20 October 2026.
  3. Peter Grünwald appointed Professor in Leiden cwi.nl. Checked 20 October 2026.
  4. Peter Grünwald cwi.nl. Checked 20 October 2026.
  5. ERC Advanced Grant for Peter Grünwald for research on a revolutionary statistical theory cwi.nl. Checked 20 October 2026.
  6. [1906.07801] Safe Testing arxiv.org. Checked 20 October 2026.
  7. Peter Grünwald named IMS Fellow for foundational contributions to statistics cwi.nl. Checked 20 October 2026.
  8. List of CWI professors cwi.nl. Checked 20 October 2026.
  9. CWI Director and Management Team cwi.nl. Checked 20 October 2026.
  10. Peter Grünwald | European Laboratory for Learning and Intelligent Systems ellis.eu. Checked 20 October 2026.