I am a linguist with a strong statistical inclination. I am also a
PhD student at the University of Amsterdam under the supervision
of Wilker Aziz.
Currently I am working on (hierarchical) Bayesian models for
machine translation and in particular word alignment. These models
require efficient inference algorithms. Together with Wilker I
have designed a fast Gibbs sampler for our word alignment model.
More recently, I have started to work with variational inference
using stochastic gradients. This work will be published soon.
Below is a list of my publications. For transparency
reasons I have decided to publish all reviews.
As I learned more about machine learning, I started to keep
notes on topics that I am interested in. These notes do not
contain any original research but explain existing statistical
modeling techniques and expand on details that cannot be found in
Together with Christian
Schaffner I have written a lecture script on
Basic Probability Theory and Statistics. We have used the script
twice for a course in Amsterdam and keep developing, taking into
account feedback from our students. The script is currently
limited to discrete probability theory. We are grateful for any
additions and corrections.
Below are some shorter notes that people working in NLP may
Most of my code can be found on github. If you can't
find what you are looking for, feel free to write me.