Peter Grünwald's Publications until 2020

For articles in Dutch, click here.

Book

• P.D. Grünwald, The Minimum Description Length Principle. MIT Press, June 2007. 570 pages.
  

Refereed and Invited Publications

2019

• Z. Mhammedi and P.D. Grünwald and B. Guedj. PAC-Bayes Unexpected Bernstein Inequality. Proceedings of NeurIPS 2019. Available on arXiv as https://arxiv.org/abs/1905.13367, 2019.
• P.D. Grünwald and T. Roos. Minimum Description Length Revisited. International Journal of Mathematics for Industry, 2019.
• J. ter Schure and P. Grünwald. Accumulation Bias in Meta-Analysis: The Need to Consider Time in Error Control. F1000Research 2019, 8:962 https://doi.org/10.12688/f1000research.19375.1.
• T. van Ommen and W. Koolen and P. Grünwald. Efficient Algorithms for Minimax Decisions under Tree-Structured Incompleteness. Proceedings ECSQARU 2019.
• P.D. Grünwald and N. Mehta A Tight Excess Risk Bound via a Unified PAC-Bayesian-Rademacher-Shtarkov-MDL Complexity. Proceedings ALT (Algorithmic Learning Theory) 2019. Extended version available on arXiv as https://arxiv.org/abs/1710.07732.

2018

• P.D. Grünwald. Safe Probability. Journal of Statistical Planning and Inference, 195, 2018, pp. 47-63. Earlier version available on arXiv as http://arxiv.org/abs/1604.01785.
• S. van der Pas and P.D. Grünwald. Almost the Best of Three Worlds: Risk, Consistency and Optional Stopping for the Switch Criterion in Single Parameter Model Selection. Statistica Sinica 28(I), pp. 229-256, January 2018. Available on arXiv as arXiv:1408.5724.
• P.D. Grünwald and R. de Heide. Invited Discussion of Using Stacking to Average Bayesian Predictive Distributions by Yao, Vehtari, Simpson and Gelman. Bayesian Analysis 13 (3), 2018.

2017

• 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. Here is a video recording of a talk I gave about this subject at the NIPS 2014 workshop From Bad Models to Good Policies. Here is a small erratum (one formula was wrong in Section 2.5)
• N. Lewis and P.D. Grünwald. Objectively combining AR5 instrumental period and paleoclimate climate sensitivity evidence. To appear in Climate Dynamics, 2017. (author's accepted manuscript; the final publication is available as read-only pdf at Springer via this link and, if you have journal access, via this link). R code that implements the method is avalaible at Nic Lewis' website.

2016

• Peter D. Grünwald. Contextuality of Misspecification and Data-Dependent Losses: Comment on Watson and Holmes (2016) In Statistical Science 31(4), pages 495- 498, December 2016.
• Wouter Koolen, Peter D. Grünwald and T. van Erven. Combining adversarial guarantees and stochastic fast rates in online learning In Advances in Neural Information Processing Systems (NIPS) 29, pages 4457-4465, December 2016. Here is a longer version containing proofs.
• Peter D. Grünwald. Toetsen als Gokken: een redelijk alternatief voor de p-waarde (in Dutch). Nieuw Archief voor Wiskunde 5/17 Nr 4., December 2016.
• Thijs van Ommen, Wouter M. Koolen, Thijs E. Feenstra and Peter D. Grünwald. Robust Probability Updating. International Journal of Approximate Reasoning, vol. 74, pp. 30-57, July 2016.
• R. Shiffrin, S. Chandramouli, and P.D. Grünwald. Bayes Factors, Relations to Minimum Description Length, and Overlapping Model Classes. Journal of Mathematical Psychology, vol 72, pp. 56-77, 2016.
• K. Oikonomou and P.D. Grünwald. Explicit Bounds for Entropy Concentration under Linear Constraints. Available on arXiv as arXiv:1107.6004. IEEE Transactions on Information Theory, vol. 62, nr. 3, pp. 1206-1230, 2016.

2015

• T. van Erven, P.D. Grünwald, N. Mehta, M. Reid and R. Williamson. Fast Rates in Statistical and Online Learning. Journal of Machine Learning Research, vol. 16, pp. 1793-1861, 2015, Special issue dedicated to the Memory of Alexey Chervonenkis, 2015.

2014

• T. van Erven and W. Koolen an P.D. Grünwald. Learning the Learning Rate for Prediction with Expert Advice. Advances in Neural Information Processing Systems 27 (NIPS 2014).
• S. de Rooij and T. van Erven and P.D. Grünwald and W. Koolen. Follow the Leader if You Can, Hedge if You Must. Journal of Machine Learning Research 15, pages 1281-1316, April 2014.
• J. Witteveen, W. Duivesteijn, A. Knobbe and P. D. Grünwald. RealKrimp - Finding Hyperintervals that Compress with MDL for Real-Valued Data. Proceedings IDA 2014. Springer Lecture Notes in Computer Science LNCS 8819, pp. 368--379.

2013

• P.D. Grünwald. Safe Probability: restricted conditioning and extended marginalization. Proceedings Twelfth European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty (ECSQARU 2013). Lecture Notes in Computer Science, Vol. 7958, Springer Verlag 2013, pages 242-253.
• P. Bartlett, P. Grünwald, P. Harremoes, F. Hedayati, W. Kotlowski. Horizon-Independent Optimal Prediction with Log-Loss in Exponential Families, JMLR Workshop and Conference Proceedings 30, COLT 2013: 662-695, 2013.

2012

• J.W. Romeijn, T. Sterkenburg and P.D. Grünwald. Good Listeners, Wise Crowds and Parasitic Experts (I). Comment on Meta-Induction and the Wisdom of Crowds by Paul D. Thorn and Gerhard Schurz. Analyse & Kritik 34(2), pages 399-408, 2012.
• T. van Erven, P.D. Grünwald, M.D. Reid and R.C. Williamson Mixability in Statistical Learning. Advances in Neural Information Processing Systems 24 (NIPS 2012) (longer version with proofs included).
• P.D. Grünwald. The Safe Bayesian: learning the learning rate via the mixability gap. Proceedings 23rd International Conference on Algorithmic Learning Theory (ALT '12), Copyright © 2012 Springer Verlag. The link is to a slightly longer version than the conference paper, including the proof of Theorem 3 and a little additional discussion.
• P.D. Grünwald. Commentary on The Optimality of Jeffreys Prior for Online Density Estimation and the Asymptotic Normality of Maximum Likelihood Estimators by F. Hedayati and P. Bartlett (I). JMLR Workshop and Conference Proceedings Volume 23, COLT 2012: 7.14 - 7.17, 2012
• T. van Erven and P.D. Grünwald and S. de Rooij. Catching Up Faster by Switching Sooner: A Predictive Approach to Adaptive Estimation with an application to the AIC-BIC Dilemma. Journal of the Royal Statistical Society, Series B 74(3), 361-397 (with discussion, pp. 397-417), 2012. Presented as a ``Read Paper'' at the RSS Ordinary Meeting 19 October 2011. Preprint. Matlab code .

2011

• P.D. Grünwald and J.Y. Halpern. Making Decisions Using Sets of Probabilities: Updating, Time Consistency, and Calibration. Journal of Artificial Intelligence Research (JAIR) 42, pages 393-426, 2011.
• T. van Erven, P.D. Grünwald , W. Koolen and S. de Rooij. Adaptive Hedge. Advances in Neural Information Processing Systems 24 (NIPS 2011), pp. 1656--1664, Granada, Spain, 2011.
• P.D. Grünwald. Safe Learning: bridging the gap between Bayes, MDL and statistical learning theory via empirical convexity. Proceedings 24th Conference on Learning Theory (COLT 2011), pp. 551--573, Budapest, 2011.
• W. Kotlowski and P.D. Grünwald. Maximum Likelihood vs. Sequential Normalized Maximum Likelihood in On-line Density Estimation. Proceedings 24th Conference on Learning Theory (COLT 2011), pp. 761-779. Budapest, 2011.
• S. de Rooij and P.D. Grünwald. Luckiness and Regret in Minimum Description Length Inference. Handbook of the Philosophy of Science, Volume 7: Philosophy of Statistics (edited by Prasanta S. Bandyopadhyay and Malcolm Forster), pages 865-900. Elsevier Science Publishers, 2011, ISBN 0444518622.

2010

• S. K. Bar-Lev, D. Bshouty, P.D. Grünwald and P. Harremoes. Jeffreys vs. Shtarkov Distributions Associated with Some Natural Exponential Families. Statistical Methodology (2010), doi:10.1016/j.stamet.2010.06.001, in press.
• P.D. Grünwald and W. Kotlowski Prequential Plug-In Codes that Achieve Optimal Redundancy Rates even if the Model is Wrong (the link is to an extended version containing proofs of the main results). Proceedings of the 2010 International Symposium on Information Theory (ISIT 2010), Houston, Texas, 2010.
• W. Kotlowski, P.D. Grünwald and S. de Rooij. Following the Flattened Leader. Proceedings 23rd Conference on Learning Theory (COLT 2010), Haifa, 2010.

2009

• P.D. Grünwald and P. Harremoes. Finiteness of Redundancy, Regret, Shtarkov Sums and Jeffreys Integrals in Exponential Families. Proceedings of the 2009 International Symposium on Information Theory (ISIT 2009), Seoul, South-Kora, 2009.
• P.D. Grünwald and D. Navarro. NML, Bayes and true distributions: A comment on Karabatsos and Walker. Journal of Mathematical Psychology 53, pages 43-51, 2009 (doi:10.1016/j.jmp.2008.11.005).

2008

• P.D. Grünwald. That Simple Device Already Used By Gauss. In Festschrift in Honor of Jorma Rissanen on the Occasion of his 75th Birthday, Tampere University Press, 2008.
• R.D. Gill and P.D. Grünwald. An Algorithmic and a Geometric Characterization of Coarsening at Random. Annals of Statistics 36(5), pages 2409-2422, 2008.
• P.D. Grünwald and J.Y. Halpern. A Game-Theoretic Analysis of Updating Sets of Probabilities . Proceedings of the Twenty-Fourth Annual Conference on Uncertainty in Artificial Intelligence (UAI 2008), pages 240--247, Helsinki, Finland, July 2008. The link is to a longer version of the paper (math/CS arXiv, arXiv:0711.3235v1, 2007), including the proofs that are left out of the conference version. Paper now superseded by our 2011 JAIR paper.
• P.D. Grünwald and P.M.B. Vitányi. Algorithmic Information Theory. In Handbook of the Philosophy of Science, Volume 8: Philosophy of Information. (edited by P. Adriaans and J. van Benthem), pp 289-325. Elsevier Science Publishers, 2008.
• P.D. Grünwald. Entropy Concentration and the Empirical Coding Game. Statistica Neerlandica 62(3), pages 374-392, 2008. Special Issue: Eurandom 1998-2008: A random tour through a decade of research. Here is a slightly modified version.
• P.D. Grünwald, P. Myllymäki, I. Tabus, M. Weinberger and B. Yu (editors). Festschrift in Honor of Jorma Rissanen on the Occasion of his 75th Birthday. Tampere University Press, 2008.

2007

• T. van Erven and P.D. Grünwald and S. de Rooij. Catching up Faster in Bayesian Model Selection and Model Averaging. Advances in Neural Information Processing Systems 20 (NIPS 2007).
• P.D. Grünwald and J. Langford. Suboptimal behavior of Bayes and MDL in classification under misspecification. Machine Learning 66(2-3), DOI 10.1007/s10994-007-0716-7, pages 119-149, 2007.
• P.D. Grünwald, The Minimum Description Length Principle. MIT Press, June 2007. 570 pages.

2006

• P.D. Grünwald. Bayesian Inconsistency under Misspecification. Four page abstract of a plenary presentation at the Valencia 8 ISBA conference on Bayesian statistics, June 2006 (the link is to a slightly extended version of the abstract).
• P.D. Grünwald. Review of the book Statistical and Inductive Inference by Minimum Message Length by Chris Wallace, Springer 2005. \(I) Computer Journal, June 2006.
• S. de Rooij and P.D. Grünwald. An Empirical Study of MDL Model Selection with Infinite Parametric Complexity. Journal of Mathematical Psychology 50(2), pages 180-192, 2006.
• E.J. Wagenmakers, P.D. Grünwald and M. Steyvers. Accumulative prediction error and the selection of time series models. Journal of Mathematical Psychology 50(2), pages 149-166, 2006.
• E.J. Wagenmakers and P.D. Grünwald. A Bayesian perspective on hypothesis testing: a Comment on Killeen (2005). Psychological Science 17(7), pages 641-642, 2006.
• P. Grünwald. A First Look at the Minimum Description Length Principle. Chapter 12 in Intelligent Algorithms in Ambient and Biomedical Computing (edited by W. Verhaegh, E. Aarts, and J. Korst), pp 187-213. Philips Research Book Series, Vol. 7, Springer-Verlag.

2005

• T. Roos, , P. Grünwald, P. Myllymäki and H.Tirri. Generalization to Unseen Cases. Advances in Neural Information Processing Systems 18 (NIPS 2005), pages 1129--1136, 2006.
• S. de Rooij and P.D. Grünwald. MDL Model Selection using the ML Plug-in Code. Proceedings of the 2005 IEEE International Symposium on Information Theory (ISIT 2005), September 2005.
• W. van Dam, R.D. Gill and P.D. Grünwald. The statistical strength of nonlocality proofs. IEEE Transactions on Information Theory 51(8), 2812-2835, August 2005.
• T. Roos, H. Wettig, P. Grünwald, P. Myllymäki and H.Tirri. On discriminative Bayesian Network classifiers and logistic regression. Machine Learning 59(3), pages 267 - 296, 2005.
• P.D. Grünwald and S. de Rooij. Asymptotic Log-Loss of Prequential Maximum Likelihood Codes. Proceedings of the Eighteenth Annual Conference on Learning Theory (COLT 2005), pages 652-667, June 2005.
• P.D. Grünwald, I.J. Myung, M. Pitt (editors). Advances in Minimum Description Length: Theory and Applications. MIT Press, 2005.
• P. Grünwald. A Tutorial Introduction to the Minimum Description Length Principle. Chapters 1 and 2 of Advances in Minimum Description Length: Theory and Applications (edited by P. Grünwald, I.J. Myung, M.A. Pitt), MIT Press, April 2005. (80 pages; postscript and PDF).
  Errata can be found here.

2004

• P.D. Grünwald and A.P. Dawid. Game theory, maximum entropy, minimum discrepancy, and robust Bayesian decision theory. Annals of Statistics 32(4), pages 1367-1433, 2004.
• P.D. Grünwald and J. Halpern. When ignorance is bliss. Proceedings of the Twentieth Annual Conference on Uncertainty in Artificial Intelligence (UAI 2004), Banff, Canada, July 2004.
• P.D. Grünwald and J. Langford. Suboptimal behaviour of Bayes and MDL in classification under misspecification. Proceedings of the Seventeenth Annual Conference on Learning Theory (COLT 2004), July 2004 (the paper on this webpage is a slight extension of the conference paper, containing two extra pages of discussion).

2003

• P.D. Grünwald and P.M.B. Vitányi. Kolmogorov complexity and information theory with an interpretation in terms of questions and answers . Journal of Logic, Language and Information 12, pages 497-529, 2003. Click here for postscript and here for pdf.
• P.D. Grünwald and J. Halpern. Updating probabilities. Journal of Artificial Intelligence Research (JAIR) 19, pages 243-278, 2003.
• H. Wettig, P. Grünwald, T. Roos, P. Myllymäki and H.Tirri. When discriminative learning of Bayesian network parameters is easy. In Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence (IJCAI 2003), pages 491-496, Acapulco, Mexico, August 2003 (postscript or pdf ).

2002

• P. Grünwald and J. Halpern. Updating probabilities. In Proceedings of the Eighteenth Annual Conference on Uncertainty in Artificial Intelligence (UAI 2002), pages 187-196, University of Alberta, Edmonton, Canada, August 2002. ( postscript or pdf)
• P. Grünwald, P. Kontkanen, P. Myllymäki, T. Roos and H. Tirri. Supervised posterior distributions. Abstract, presented at the 7th Valencia International Meeting on Bayesian Statistics, Tenerife, Spain, June 2002.
• P. Grünwald. Taking the sting out of subjective probability In Words, Proofs, and Diagrams (editors D. Barker-Plummer, D. Beaver, J. van Benthem and P. Scotto Di Luzio), pages 75--94. CSLI Publications, Stanford, CA, June 2002.

2001

• P. Grünwald. Strong entropy concentration, game theory and algorithmic randomness. In Proceedings of the Fourteenth Annual Conference on Computational Learning Theory (COLT 2001), pages 320-336. Amsterdam, the Netherlands, July 2001. Copyright © 2001 Springer Verlag.

2000

• P.D. Grünwald. Model selection based on minimum description length, Journal of Mathematical Psychology 44, pages 133-152, 2000. (if you are interested in learning about MDL, please refer to my much more recent Minimum Description Length Tutorial cited above; the JMP article is somewhat dated).
• P.Kontkanen, P. Myllymäki, T. Silander, H.Tirri, and P.D. Grünwald. Predictive distributions and Bayesian networks, Journal of Statistics and Computing 10, pages 39-54, 2000.
• P. Grünwald. Maximum entropy and the glasses you are looking through. In Proceedings of the Sixteenth Annual Conference on Uncertainty in Artificial Intelligence (UAI 2000), pages 238-246. Stanford, CA, USA, July 2000.

1999

• P. Grünwald. Viewing all models as 'probabilistic'. In Proceedings of the Twelfth Annual Conference on Computational Learning Theory (COLT' 99), pages 171-182. Santa Cruz, CA, USA, July 1999.

1998

• P. Grünwald, P.Kontkanen, P. Myllymäki, T. Silander and H.Tirri. Minimum encoding approaches for predictive modeling. In Proceedings of the Fourteenth International Conference on Uncertainty in Artificial Intelligence (UAI'98), pages 183-192. Madison, WI, USA, July 1998.
• P.Kontkanen, P. Myllymäki, T. Silander, H.Tirri, and P. Grünwald. Bayesian and information-theoretic priors for Bayesian network parameters, pages 89-94 in Machine Learning: ECML-98, Proceedings of the 10th European Conference, edited by C.Nédellec and C.Rouveirol, pages 89-94. Vol. 1398 in Lecture Notes in Artificial Intelligence, Springer-Verlag, 1998.
• P.Kontkanen, P.Myllymäki, T.Silander, H.Tirri, and P. Grünwald. On the small sample size behavior of Bayesian and information-theoretic approaches for predictive inference. Abstract, presented at the 6th Valencia International Meeting on Bayesian Statistics, Alcossebre, Spain, May-June 1998.
• P. Grünwald. Ramifications and sufficient causes. Abstract, presented at the Fourth Symposium on Logical Formalizations of Common Sense Reasoning (Common Sense '98), London 1998.

1997

• P.Kontkanen, P. Myllymäki, T. Silander, H.Tirri, and P. Grünwald. Comparing predictive inference methods for discrete domains. In Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics (AISTATS '97), pages 311-318. Fort Lauderdale, Florida, 1997.
• P. Grünwald. Causation and nonmonotonic temporal reasoning. In KI-97: Advances in Artificial Intelligence (editors G. Brewka, C. Habel and B. Nebel), pages 159-170; Lecture Notes in Artificial Intelligence no. 1303. Springer Verlag, Berlin, Germany, 1997.

1996

• M. Steyvers and P. Grünwald. A recurrent network that performs a context-sensitive prediction task. In Proceedings Eighteenth Annual Conference of the Cognitive Science Society, pages 335-339. Morgan Kauffman, 1996.
• P. Grünwald. A minimum description length approach to grammar inference In Symbolic, Connectionist and Statistical Approaches to Learning for Natural Language Processing (editors S. Wermter, E. Riloff, G. Scheler), pages 203-216; Lecture Notes in Artificial Intelligence no. 1040. Springer Verlag, Berlin, Germany, 1996.

Nonrefereed Publications and Publications at Local Meetings

Articles for Popular Magazines (in Dutch)

• P. Grünwald. Kansloos: van Willem Ruis tot Lucia de Berk (in Dutch). Deel I (Bi-logical 1(2), December 2008). Deel II (Bi-logical 2(1), Juni 2009).

Invited Paper

• P. Grünwald and A.P. Dawid. Game theory, maximum generalized entropy, minimum discrepancy, robust Bayes and Pythagoras, Proceedings 2002 Information Theory Workshop (ITW 2002), pages 94-97. Bangalore, India, October 2002 (Click here for postscript and here for pdf).

Ph.D. Thesis

P. Grünwald. The Minimum Description Length Principle and Reasoning under Uncertainty, ILLC Dissertation series 1998-03, University of Amsterdam, 1998. Click here.

Technical Reports

• P. Grünwald. Strong entropy concentration, game theory, coding and randomness, EURANDOM Technical Report 2001-010, Eindhoven, 2001.

Publications at Workshops and Local Conferences

• T. van Erven and S. de Rooij and Peter Grünwald. Switching between Predictors with an Application in Density Estimation. Proceedings of the 28th Symposium on Information Theory in the Benelux, Enschede, The Netherlands, 2007.
• T. Roos, , P. Grünwald, P. Myllymäki and H.Tirri. Generalization to Unseen Cases. In Proceedings BNAIC '05 (Belgium-Netherlands conference on Artificial Intelligence), Bruxelles, Belgium, October 2005 Winner of Best Paper Award!
• P. Grünwald and J. Halpern. Updating probabilities (abstract). In Proceedings BNAIC '03 (Belgium-Netherlands conference on Artificial Intelligence) , Nijmegen, the Netherlands, October 2003.
• H. Wettig, P. Grünwald, T. Roos, P. Myllymäki and H. Tirri. Supervised learning of Bayesian network parameters made easy. In Proceedings BeNeLearn '02 (Belgium-Netherlands conference on Machine Learning), Utrecht, the Netherlands, 2002.
• P.Kontkanen, P. Myllymäki, T. Silander, H.Tirri, and P. Grünwald. On predictive distributions and Bayesian networks. In Proceedings BeNeLearn '97 (Belgium-Netherlands conference on Machine Learning), Tilburg, the Netherlands, 1997.
• P.Kontkanen, P. Myllymäki, T. Silander, H.Tirri, and P. Grünwald. Comparing predictive inference methods for discrete domains. Pages 311-318 in Proceedings of the Sixth International Workshop on Artificial Intelligence and Statistics (Ft. Lauderdale, USA, January 1997).
• P. Grünwald. The minimum description length principle and non-deductive inference. Proceedings IJCAI Workshop on Abduction and Induction in AI (editor P. Flach), Nagoya, Japan 1997.
• P. Grünwald. Nonmonotonic temporal reasoning as a search for explanations. NRAC '97 (Second IJCAI Workshop on Nonmonotonic Reasoning, Action and Change), Nagoya, Japan, 1997.
• P. Grünwald. Causation, explanation and persistence. In Proceedings 1997 Dutch German Workshop on Nonmonotonic Reasoning, pages 149-158, Saarbrücken 1997.
• P. Grünwald. Causal networks and nonmonotonic temporal reasoning. In Proceedings 1996 Dutch Conference on Artificial Intelligence (NAIC-96), pages 157-166, nominated for Best Paper Award, Utrecht 1996.
• P. Grünwald, B. Gaume and M. Bouajjani. A new causal theory of action. In Proceedings 1995 Dutch Conference on Artificial Intelligence (NAIC-95), Rotterdam 1995.