I am a Senior Research Scientist in Pete Battaglia’s Structured Intelligence Team at DeepMind, London.
My current research focuses on learning physical simulations of the world we inhabit. A key challenge I obsess about is how to build high precision machine learning models, with controllable error down to physics simulator accuracy and exhibiting the gurantees and properties of classical numerical methods, while leveraging the power of data-driven techniques.
More generally I am interested in AI4Science, equivariance, and combinatorial optimization, while previously I have worked on uncertainty quantification, unsupervised representation learning, variational inference, normalizing flows, optimization, and medical imaging.
I am proud contributor to the UvA Inclusive AI Initiative, which seeks to promote diversity and inclusion in the academic AI world.
- July 2022: I will be lecturing at PALMS 2022 Machine Learning Summer School.
- April 2022: I joined the Structured Intelligence Team at DeepMind as a Senior Research Scientist.
- March 2021: I talked at ELLIS Symposium Workshop on Geometric Deep Learning for Medical Imaging on Equivariance - Trends and Challenges. Watch the video
- February 2021: I talked at IBM Research on Exploiting Symmetries for Deep Learning
- September 2020: I joined as a Staff Engineer at Qualcomm AI Research, Amsterdam.
- August 2020: I lectured at the Machine Learning Summer School-Indonesia.
- September 2018: I have begun lecturing a new course at the UvA on Probability Theory for Machine Learning
- June 2018: I talked at London Machine Learning Meetup on Understanding & Generalising the Convolution. Watch the video
- UvA AMLAB Seminars
- UvA Geometric Methods: Riemannian Geometry Notes
- UCL Computer Vision Reading Group
2020 - 2022: Staff Engineer working on Machine Learning for Combinatorial Optimization at Qualcomm AI Research, Amsterdam.
2018 - 2020: Postdoc in Amsterdam Machine Learning Lab (AMLAB) with Max Welling at the University of Amsterdam. Concurrently lectured a fresh course in Bayesian Statistics for Machine Learning, and was lab manager for the Philips lab at the UvA.
2014 - 2018: PhD in Computer Vision at UCL with Gabriel J. Brostow and Dr. Clare Wilson FRCOphth.
2010 - 2014: Engineering at The University of Cambridge (I am a scholar of Sidney Sussex College).
Previous Masters Students
Alessandra van Ree: Neural PDE Solvers
Eli de Smet: Simple Saddle-Free Newton Descent with a Pseudo-Hessian
Jan Huiskes: FF-net: A fast Fourier transform inspired neural network
Rob Hesselink: Contrastive Learning of Equivariant Representations
Vivian van Oijen: Esophageal histopathology screening
Bryan Cardenas: Esophageal histopathology screening
Christina Winkler: Conditional Normalizing Flows
Nichita Diaconu: Learning convolutions
Jonas M J Braun: Non-geometric equivariance
Marios Fournarakis: Equivariance for tracking
Tycho van der Ouderaa: Reversible Networks for Memory-efficient Image-to-Image Translation in 3D Medical Imaging
Saki Shinoda: Semi-supervised learning