Pierre Humbert

Postdoc in probability and statistics

 

Since September 2024, I am a postdoctoral researcher in the Laboratoire de Probabilités, Statistique et Modélisation (LPSM) at Sorbonne Université with Etienne Roquain. I am also a membre of the MARS project.

Prior to that, I was a postdoc in the probability and statistics team of the Laboratoire de Mathématiques d'Orsay (LMO) with Sylvain Arlot. I was also a member of the INRIA Celeste team. I completed a PhD at ENS Paris-Saclay in 2021 entitled Multivariate analysis with tensors and graphs - application to neuroscience, under the supervision of Nicolas Vayatis, Laurent Oudre, and Julien Audiffren.

 

See my Google Scholar.

CV is available here .

 

Main research interests

  • Conformal prediction
  • Statistical learning, non-parametric statistics
  • Cross-validation, robust statistics
  • Signal processing
  • Graph and tensor learning
  • Applications to neuroscience

Contact

prenom.nom@sorbonne-universite.fr

Main publications

Journals

  • (2024) L. Zoroddu, P. Humbert, L. Oudre Learning Network Granger causality using Graph Prior Knowledge
    Transactions on Machine Learning Research (TMLR). [pdf]

  • (2021) P. Humbert*, B. Le Bars*, L. Oudre, A. Kalogeratos, N. Vayatis. Learning Laplacian matrix from graph signals with sparse spectral representation
    Journal of Machine Learning Research (JMLR). [journal] [pdf] [code]

  • (2021) P. Humbert, L. Oudre, N. Vayatis, J. Audiffren. Tensor convolutional dictionary learning with CP low-rank activations
    IEEE Transactions on Signal Processing (TSP). [journal] [pdf] [code]

  • (2019) P. Humbert, C. Dubost, J. Audiffren, L. Oudre. Apprenticeship learning for a predictive state representation of anesthesia
    IEEE Transactions on Biomedical Engineering (TBME). [journal] [pdf]

 

Conferences

  • (2023) P. Humbert, B. Le Bars, A. Bellet, S. Arlot. One-shot federated conformal prediction
    International Conference on Machine Learning (ICML), 2023. [hal] [pdf] [code]
  • (2022) P. Humbert*, B. Le Bars*, L. Minvielle*. Robust kernel density estimation with median-of-means principle
    International Conference on Machine Learning (ICML), 2022. [conference] [pdf] [code]

  • (2021) P. Humbert, L. Oudre, C. Dubost. Learning spatial filters from EEG signals with graph signal processing methods
    IEEE Engineering in Medecine and Biology Society (EMBC), 2021. [pdf]

  • (2021) T. Gnassounou, P. Humbert, L. Oudre. Adaptive subsampling of multidomain signals with graph products
    International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2021. [conference] [pdf] [code]

  • (2020) B. Le Bars, P. Humbert, A. Kalogeratos, N. Vayatis. Learning the piece-wise constant graph structure of a varying Ising model
    International Conference on Machine Learning (ICML), 2020. [conference] [pdf] [code]

  • (2020) P. Humbert, J. Audiffren, L. Oudre, N. Vayatis. Low rank activations for tensor-based convolutional sparse coding
    International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2020. [conference] [pdf] [code]

  • (2019) P. Humbert, L. Oudre, N. Vayatis. Subsampling of multivariate time-vertex graph signals
    European Signal Processing Conference (EUSIPCO), 2019. [conference]

  • (2019) B. Le Bars*, P. Humbert*, L. Oudre, A. Kalogeratos. Learning Laplacian matrix from bandlimited graph signals
    International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2019. [conference] [pdf] [code]

 

Preprints

  • (2024) P. Humbert, B. Le Bars, A. Bellet, S. Arlot. Marginal and training-conditional guarantees in one-shot federated conformal prediction [hal] [pdf]

  • (2024) J-B. Fermanian, P. Humbert. Conditional Prediction Sets with Weighted Conformal Prediction