I am a research associate at the University of Amsterdam working on natural language processing problems such as machine translation, word alignment, textual entailment, and paraphrasing. My interests sit at the intersection of disciplines such as formal languages, machine learning, approximate inference, global optimisation, and computational linguistics.
Recently, I’ve developed quite an interest in Bayesian deep learning. In particular, I’m developing probabilistic neural network models that reason with and induce forms of discrete generalisation such as trees and graphs.
If you are looking for a project with me, you might want to start by joining my reading group on approximate inference. I’ve put together a list of what I think may help navigate through the landscape of deep generative models. And finally, I’ve made some technical notes available on github.
If you need to find me try Science Park 107 (F2.11).
- May 21 We will present our work on modelling latent variation in translation data at ACL18 (links to code and paper coming soon!).
- I am co-organising a DGM day at UvA where we will be presenting our VI tutorial as well as a bunch of interesting DGMs in NLP. Join us on March 22, 2018!
- Philip and I are going to present our tutorial on variational inference and deep generative models at ACL 2018! See you in Melbourne!
- We will present our work on generative models of joint word representation and alignment at NAACL18 (code and paper).