Hi! I am Tom, I am a Sofware Engineer at Facebook and I used to be a PhD student in the DTAI Research group at KU Leuven working with Prof. Jesse Davis.
Download my resume.
Publications
For the most up-to-date list, see google scholar.
PhD Thesis
- Soccer Analytics Meets Artificial Intelligence: Learning Value and Style from Soccer Event Stream Data. Download thesis.
Conferences
- SoccerMix: Representing Soccer Actions with Mixture Models Tom Decroos, Maaike Van Roy, Jesse Davis. ECML/PKDD 2020. paper slides
- Actions Speak Louder Than Goals: Valuing Player Actions in Soccer. Tom Decroos, Lotte Bransen, Jan Van Haaren, Jesse Davis. KDD 2019 (Best Applied Data Science Paper). paper slides youtube
- Player Vectors: Characterizing Soccer Players’ Playing Style from Match Event Streams. Tom Decroos, Jesse Davis. ECML/PKDD 2019. paper slides
- Automatic Discovery of Tactics in Spatio-Temporal Soccer Match Data.
Tom Decroos, Jan Van Haaren, Jesse Davis. KDD 2018. paper slides poster
- AMIE: Automatic Monitoring of Indoor Exercises. Tom Decroos, Kurt Schütte, Tim Op De Beéck, Benedicte Vanwanseele, Jesse Davis. ECML/PKDD 2018. paper slides poster
- Predicting Soccer Highlights from Spatio-Temporal Match Event Streams. Tom Decroos, Vladimir Dzyuba, Jan Van Haaren, Jesse Davis. AAAI 2017. paper poster
Workshops
- Interpretable Prediction of Goals in Soccer. Tom Decroos, Jesse Davis. Team Sports in AI workshop at AAAI 2020 / StatsBomb Innovation in Football Conference. paper slides youtube
- Valuing On-the-Ball Actions in Soccer: A Critical Comparison of xT and VAEP. Maaike Van Roy, Pieter Robberechts, Tom Decroos, Jesse Davis. Team Sports in AI workshop at AAAI 2020. paper slides.
- Analyzing Soccer Players’ Skill Ratings Over Time Using Tensor-Based Methods. Kenneth Verstraete, Tom Decroos, Bruno Coussement, Nick Vannieuwenhoven, Jesse Davis. Machine Learning and Data Mining for Sports Analytics ECML/PKDD 2019 workshop. paper
- Characterizing Soccer Players’ Playing Style from Match Event Streams. Aron Geerts, Tom Decroos, Jesse Davis. Machine Learning and Data Mining for Sports Analytics ECML/PKDD 2018 workshop. paper slides
- STARSS: A spatio-temporal action rating system for soccer. Tom Decroos, Jan Van Haaren, Vladimir Dzyuba, Jesse Davis. Machine Learning and Data Mining for Sports Analytics ECML/PKDD 2017 workshop. paper slides
- Predicting the potential of professional soccer players. Ruben Vroonen, Tom Decroos, Jan Van Haaren, Jesse Davis. Machine Learning and Data Mining for Sports Analytics ECML/PKDD 2017 workshop. paper slides
Media
Our work on valuing actions in soccer appeared in numerous newspapers and media outlets:
Software
I am the author and maintainer of two soccer-related pip packages:
-
socceraction
https://github.com/ML-KULeuven/socceraction: Convert soccer event stream data from commercial vendors (e.g., Opta, Wyscout, StatsBomb) to actions in the simpler SPADL language and value the actions using the VAEP framework.
-
matplotsoccer
https://github.com/TomDecroos/matplotsoccer: Visualize soccer event stream data and common figures such as heatmaps over a soccer field.
Reviewing
Organizing
I co-organized a three day Spring workshop on mining and learning (SMiLee) for the ML group of KU Leuven in 2017.
Didactical tasks
I was a teaching assistant for the following courses:
I was a daily master’s thesis advisor for the following students:
- 2019-2020: Jasper Maes, Tomas Geens, James Defauw
- 2018-2019: Tim Hofmans, Nick Schouten
- 2017-2018: Kenneth Verstraete, Wouter Bruyninckx, Aron Geerts
- 2016-2017: Ruben Vroonen
I was the ombuds for both the Dutch Master Computerwetenschappen and the English Master in Computer Science at KU Leuven.
Email
<firstname>.<lastname>@gmail.com
You can also reach me via my LinkedIn or Twitter.
Last updated: 06/10/2020