Lisa Schut

Photo credit: Alina l'Ami

About Me

I'm currently a Research Assistant in the OATML lab at the University of Oxford, under the supervision of Yarin Gal. I'm curious about a wide range of topics, generally falling under the areas of statistics, machine learning, and optimization. Currently, my focus is on understanding machine learning algorithms and ensuring that they are implemented robustly.

My undergraduate degree is in Econometrics and Operations Research from Erasmus University Rotterdam, and I have two masters in the fields of Statistics and Computer Science from the University of Oxford.

Since the age of eight, I've been passionate about playing chess. Originally, chess was simply a family activity, but incipiently I became enamored with the game - I love the interplay between creativity, sharp analytical skills, and psychology. At the age of twelve, I played my first World Youth Championship and by the age of fourteen, joined the Dutch Olympic Chess team. In 2014, we founded ChessQueens to improve gender equality in chess in the Netherlands, by supporting female chess players.

I'm excited to start my DPhil coming fall, as a member of the AIMS CDT and OATML lab.

Publications

Capsule Networks -- A Probabilistic Perspective. Lewis Smith, Lisa Schut, Yarin Gal and Mark van der Wilk. Spotlight paper at the ICML Workshop on Object-Oriented Learning: Perception, Representation, and Reasoning, 2020. [Workshop Paper] [ArXiv pre-print]

Influence of Outliers on Cluster Correspondence Analysis. Michel van de Velden, Alfonso Iodice D'Enza, Lisa Schut. Accepted for talk at CLADAG, Cassino Italy 2019. [Conference proceedings]

An Investigation into Different Architectures for Learning Word Representations. Lisa Schut, Adam Hare, Hans Hanley and Aditya Agarwal. Accepted as poster to OxCSC, Oxford United Kingdom 2019.

Reviewing

  • ICML Workshop in Human Interpretability 2020

Contact

If you're interested in my CV or want to reach out, feel free to drop me a line at lisa dot schut @ cs dot ox dot ac dot uk.