Peter Grünwald heads the machine learning group at CWI in Amsterdam, the Netherlands. He is also full professor of Statistical Learning at the mathematical institute of Leiden University. He regularly publishes at the world’s top machine learning venues such as NIPS, UAI and COLT and in top statistics journals such as the Annals of Statistics. He has been co-program chair of the UAI (2010) and COLT (2015) conferences , and has been 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. In 2010 he was co-awarded the Van Dantzig prize (2010), the highest Dutch award in statistics and operations research. He received NWO VIDI (2005), VICI (2010) and TOP-1 (2016) grants.
His research currently focuses on Safety and Luckiness. The basic idea is to make sure that inference from data is done in – indeed – a safer way! Most recent publications in this direction:
- P.D. Grünwald and T. van Ommen. Inconsistency of Bayesian Inference for Misspecified Linear Models, and a Proposal for Repairing It . Bayesian Analysis, 2017, pp. 1069-1103.
- P.D. Grünwald. Safe Probability. Journal of Statistical Planning and Inference, 2017. See also http://arxiv.org/abs/1604.01785.
- P.D. Grünwald and N. Mehta, 2017a (arXiv preprint). A Tight Excess Risk Bound via a Unified PAC-Bayesian-Rademacher-Shtarkov-MDL Complexity.
- P.D. Grünwald and N. Mehta, 2017b (arXiv preprint). Fast Rates for General Unbounded Loss Functions: from ERM to Generalized Bayes.
- P.D. Grünwald and R. de Heide. Why Optional Stopping is a problem for Bayesians, 2017.