skip to content

Faculty of Mathematics




I am doing a PhD in the Department of Applied Mathematics and Theoretical Physics (DAMTP) as part of Machine Learning and AI for Medicine group led by Prof. Mihaela van der Schaar. The purpose of the thesis is to develop new interpretability methods for Machine Learning models. The notion of interpretability has become central in Machine Learning since the large-scale deployment of Machine Learning models requires trust, which relies heavily on the ability of human beings to understand the predictions of such models.



  • 2020 -  Now : PhD in Machine Learning at DAMTP
  • 2019 - 2020 : Research in Theoretical Physics at ULB
  • 2018 - 2019 : Part III in Theoretical Physics at DAMTP (distinction)
  • 2017 - 2018 : M1 in Physics at ENS Paris (mention bien)
  • 2014 - 2017 : Bachelor in Engineering at ULB (la plus grande distinction)


Selected publications

  • Crabbé, J., Qian Z., Imrie F., & van der Schaar, M. (2021). Explaining Latent Representations with a Corpus of Examples. In Advances in Neural Information Processing Systems. [Paper]
  • Crabbé, J., van der Schaar, M. (2021). Explaining Time Series Predictions with Dynamic Masks. In Proceedings of the 38th International Conference on Machine Learning, PMLR 139:2166-2177. [Paper] [Video]
  • Crabbé, J., Zame, W. R., Zhang, Y., & van der Schaar, M. (2020). Learning outside the black-box: the pursuit of interpretable models. In H. Larochelle, M. Ranzato, R. Hadsell, M. F. Balcan, & H. Lin (Eds.), Advances in Neural Information Processing Systems (pp. 17838--17849). Curran Associates, Inc. [Paper] [Video]

Research Group

Machine Learning and Artificial Intelligence




01223 760417