Rianne de Heide

Machine Learning Group, CWI
Science Park 123, Amsterdam
heide at cwi dot nl

Statistical Science Group, Mathematical Institute - Leiden University

I am a PhD student at CWI under supervision of Peter Grünwald. One day a week I work at Leiden University with Jacqueline Meulman as coordinator of the MSc programme Statistical Science. I work on problems and solutions in the intersection of machine learning and statistics where the model or data collection process is misspecified. At the moment I'm working on a general framework for hypothesis testing with Wouter Koolen and Peter Grünwald, Bayesian inference under model misspecification with Peter Grünwald, and foundations of Bayesian inference with Tom Sterkenburg. I'm also interested in the mathematical and philosophical foundations of Bayesianism, machine learning, statistics and probability theory.

Writings

Why optional stopping is a problem for Bayesians
Rianne de Heide and Peter Grünwald
arXiv 1708.08278, 2018

The Safe-Bayesian Lasso (MSc Thesis)
Rianne de Heide, 2016

Teaching

Spring 2018 I am TA for Peter Grünwald's Information Theoretic Learning course, for which I also give two lectures on online learning and bandits.

Autumn 2017 I lectured the new MasterMath course Machine Learning Theory together with Wouter Koolen and Peter Grünwald. I did the first part of the course on Statistical Learning Theory.

Spring 2017 I was TA for Peter Grünwald's Information Theoretic Learning course, and I gave a lecture on online learning.



Background
I obtained a BSc in Mathematics from the University of Groningen and a MSc in Mathematics cum laude from Leiden University, with my thesis on the Safe-Bayesian Lasso. I also hold a Bachelor of Music from the Prins Claus Conservatoire and a Master of Music from the Royal Conservatory of The Hague, both in Classical music - Horn.

Other
Software
R Package SafeBayes
Distractions
I like cycling, learning languages and classical singing. I'm a contralto, and sing in the VU-Kamerkoor and many side-projects.

CWI DISCLAIMER