Ambroise Heurtebise

I received my PhD in machine learning and statistical signal processing from Inria Saclay (Université Paris-Saclay), working in the MIND team under the supervision of Alexandre Gramfort and Pierre Ablin.

Before my PhD, I completed a Master’s degree in Data Science at the Institut Polytechnique de Paris, following a four-year undergraduate and master-level program in Mathematics at Université Paris-Dauphine.

Email  /  GitHub  /  Google Scholar  /  LinkedIn

profile photo

Research

My research focuses on multi-view learning, independent component analysis (ICA), and causal inference, as well as on learning useful representations of brain activity.

project image

Multi-View Causal Discovery without Non-Gaussianity: Identifiability and Algorithms


Ambroise Heurtebise, Omar Chehab, Pierre Ablin, Alexandre Gramfort, Aapo Hyvärinen
arXiv, 2025
arxiv / code /

We learn a causal ordering (Directed Acyclic Graph) between random variables collected from different environments, such as brain regions.
project image

MVICAD2: Multi-View Independent Component Analysis with Delays and Dilations


Ambroise Heurtebise, Omar Chehab, Pierre Ablin, Alexandre Gramfort
IEEE Transactions on Biomedical Engineering (TBME), 2025
arxiv / code /

We extend the time variability of MVICAD by allowing sources to also differ by time dilations, which is for example useful when a subject performs an auditory task.
project image

MVICAD: Multi-View Independent Component Analysis with Delays


Ambroise Heurtebise, Pierre Ablin, Alexandre Gramfort
IEEE workshop on Machine Learning for Signal Processing (MLSP), 2023
arxiv / code /

Independent Component Analysis (ICA) is a popular algorithm for learning a representation of data. We propose a version that handles data collected from different contexts, and whose representations differ only by temporal delays.




Design and source code from Jon Barron's website