Office: FC14

Artificial Intelligence Group

Department of Computer Science and Technology

University of Cambridge

15 JJ Thomson Avenue

Cambridge CB3 0FD

United Kingdom

Hi, thanks for visiting my website! My name is Agnieszka (either /aɡˈɲɛʂ.ka/ or Aga works).

I am a second-year PhD student within the Artificial Intelligence Group at the University of Cambridge and a (long-term) research intern at Mila. I am affiliated with Lucy Cavendish College. My PhD supervisors are Sean Holden and Mateja Jamnik. At Mila I am supervised by William L. Hamilton, Anirudh Goyal and Yoshua Bengio.

Broadly, I am interested in systematic and compositional generalisation in deep learning. In my PhD, I am investigating these problems in the domains of multi-agent communication and visual reasoning.

I do my best to contribute to the departmental activities as well as to the broad machine learning community (most recently, as a reviewer for ICML 2020, EMNLP 2020 and NeurIPS 2020). I also have a great passion for teaching and I currently teach and supervise undergraduate students in STEM at the AI4GoodLab 2020. For more academic activites, please refer to Varia.

I hold an MSc with Distinction in Operational Research with Data Science from the University of Edinburgh. In my MSc thesis I proposed a new approach to relational reasoning with convolutional neural networks. I also hold a BSc degree in Computer Science from the Jagiellonian University. In my BSc thesis I investigated probability distributions defining random projections in extreme learning machines under supervision of Wojciech Czarnecki.

In the meantime, I had an opportunity to intern in data science and software engineering at IBM, Architech and Barclays UK.

I’m a traveller at heart: I visited 35 countries and lived in five so far. On that note, I had a chance to learn four foreign languages, which allows me to throw awkward sentences in attempts at icebreaking to this day. I also love mountains and Andrej Ivašković.

Contact: agnieszka [dot] slowik [at] cl [dot] cam [dot] ac [dot] uk