Follow me on Twitter: @tychovdo !

Mail: tychovdo [at] gmail [dot] com

View My LinkedIn Profile

View My GitHub Profile

Tycho van der Ouderaa

Tycho is a machine learning researcher.

He is particularly interested in Generative Modeling, Computer Vision, Bayesian Deep Learning, Information Theory, Multivariate Statistics and Reinforcement Learning. Furthermore, he is excited about applying machine learning to real-world problems, such as in Medical Image Analysis and Processing. Previously, Msc AI student at University of Amsterdam, Visiting Researcher at the DIAG group of Radboud UMC Nijmegen and R&D Engineer at medical AI startup Quantib.

In April 2021, he started a PhD in Machine Learning and Artificial Intelligence at Imperial College London, supervised by Dr. Mark van der Wilk.


Learning Invariant Weights in Neural Networks

Tycho F.A. van der Ouderaa, Mark van der Wilk
In ICML 2021 (UDL Workshop). Work in progress (arXiv/code soon).
Deep Group-wise Variational Diffeomorphic Image Registration

Tycho F.A. van der Ouderaa, Ivana Isgum, Wouter B. Veldhuis, Bob D. de Vos
In MICCAI 2020 (Extended Oral at Thoracic Image Analysis Workshop)
Chest CT Super-resolution and Domain-adaptation using Memory-efficient 3D Reversible GANs

Tycho F.A. van der Ouderaa, Daniel E. Worrall, Bram van Ginneken
In MIDL 2019
Reversible GANs for Memory-efficient Image-to-Image Translation

Tycho F.A. van der Ouderaa, Daniel E. Worrall
In CVPR 2019


Reversible Networks for Memory-efficient Image-to-Image Translation in 3D Medical Imaging

This master thesis proposes a novel image-to-image translation model that adapts reversible neural networks to perform memory-efficient image-to-image translation on both paired and unpaired data. We show that using the model we are able to perform several medical imaging tasks in 3D that would otherwise have not been possible due to GPU memory constraints. The thesis was assesed by Prof. Dr. Max Welling (UvA) and supervised by Prof. Dr. Bram van Ginneken (Radboud UMC) and Daniel E.Worrall (UvA).

The code is available online at Github as well as presentation slides.
Deep Reinforcement Learning in Pac-man

This bachelor thesis explores how a computer is able to learn how to play the game of Pac-man, solely by exploring different actions and observing the patterns in the observed states using deep neural networks. The thesis showed that a technique known as Deep Q-Learning can effectively be used for this problem.

The code is available online on Github and often used by other people for their projects.


Understanding Character-level RNN-LMs in PyTorch

Developed a visualization tool for character-level RNNs to better understand language models.
Geo-Vec Word Embeddings

Evaluated the embedding performance of variational graph auto-encoders as document representation model.
Waar is mijn fiets (Where Is My Bike)

The app "Waar is mijn fiets" (Where Is My Bike) assists iOs users in keeping track of the location of their parked bikes, by taking a picture of their parked bike and storing it together with the GPS location. The app has a strong focus on friendly user-interaction and is useful in the huge bike parking lots typically found in the city of Amsterdam.

The app was downloaded thousands of times through the AppStore and was featured on iCulture and the University of Amsterdam newspaper (Folia) and media channels (Facebook).
Rubiks Cube Solver

The app "Rubiks Cube Solver"‚Äč is an android app that allows users to scan a Rubiks Cube in real-life using the smartphone camera and then assists the user in solving the cube in a minimal amount of steps using an interactive 3D interface.

The code is available online at Github, as well as an explanatory poster.