no code implementations • 17 Jan 2022 • Santiago Mazuelas, Mauricio Romero, Peter Grünwald
Supervised classification techniques use training samples to learn a classification rule with small expected 0-1 loss (error probability).
no code implementations • 17 Jun 2021 • Peter Grünwald, Thomas Steinke, Lydia Zakynthinou
We give a novel, unified derivation of conditional PAC-Bayesian and mutual information (MI) generalization bounds.
1 code implementation • 4 Jun 2021 • Rosanne Turner, Alexander Ly, Peter Grünwald
We develop E-variables for testing whether two or more data streams come from the same source or not, and more generally, whether the difference between the sources is larger than some minimal effect size.
2 code implementations • 25 Mar 2021 • Hugo Manuel Proença, Peter Grünwald, Thomas Bäck, Matthijs van Leeuwen
This novel model class allows us to formalise the problem of optimal robust subgroup discovery using the Minimum Description Length (MDL) principle, where we resort to optimal Normalised Maximum Likelihood and Bayesian encodings for nominal and numeric targets, respectively.
3 code implementations • 16 Jun 2020 • Hugo M. Proença, Peter Grünwald, Thomas Bäck, Matthijs van Leeuwen
We propose a dispersion-aware problem formulation for subgroup set discovery that is based on the minimum description length (MDL) principle and subgroup lists.
no code implementations • 21 Oct 2019 • Rianne de Heide, Alisa Kirichenko, Nishant Mehta, Peter Grünwald
We study generalized Bayesian inference under misspecification, i. e. when the model is 'wrong but useful'.
no code implementations • 21 Aug 2019 • Peter Grünwald, Teemu Roos
This is an up-to-date introduction to and overview of the Minimum Description Length (MDL) Principle, a theory of inductive inference that can be applied to general problems in statistics, machine learning and pattern recognition.
1 code implementation • 18 Jun 2019 • Peter Grünwald, Rianne de Heide, Wouter Koolen
We develop the theory of hypothesis testing based on the e-value, a notion of evidence that, unlike the p-value, allows for effortlessly combining results from several studies in the common scenario where the decision to perform a new study may depend on previous outcomes.
no code implementations • 24 Jul 2018 • Allard Hendriksen, Rianne de Heide, Peter Grünwald
It is often claimed that Bayesian methods, in particular Bayes factor methods for hypothesis testing, can deal with optional stopping.
no code implementations • NeurIPS 2016 • Wouter M. Koolen, Peter Grünwald, Tim van Erven
We consider online learning algorithms that guarantee worst-case regret rates in adversarial environments (so they can be deployed safely and will perform robustly), yet adapt optimally to favorable stochastic environments (so they will perform well in a variety of settings of practical importance).
no code implementations • 6 Apr 2016 • Peter Grünwald
We formalize the idea of probability distributions that lead to reliable predictions about some, but not all aspects of a domain.
no code implementations • NeurIPS 2014 • Wouter M. Koolen, Tim van Erven, Peter Grünwald
Most standard algorithms for prediction with expert advice depend on a parameter called the learning rate.
no code implementations • NeurIPS 2012 • Tim V. Erven, Peter Grünwald, Mark D. Reid, Robert C. Williamson
We show that, in the special case of log-loss, stochastic mixability reduces to a well-known (but usually unnamed) martingale condition, which is used in existing convergence theorems for minimum description length and Bayesian inference.
no code implementations • NeurIPS 2011 • Tim V. Erven, Wouter M. Koolen, Steven D. Rooij, Peter Grünwald
In most previous analyses the learning rate was carefully tuned to obtain optimal worst-case performance, leading to suboptimal performance on easy instances, for example when there exists an action that is significantly better than all others.