Pierre Marion

I am a PhD student in Machine Learning at Sorbonne University, in the Statistics Department (LPSM). I work under the supervision of Gérard Biau and Jean-Philippe Vert.

I am also a civil servant in Corps des Mines and a Google PhD fellow.

My current research interests regard the theory of deep learning, in particular in the large-depth limit. For more information on my research activity, you can visit the Research page.

To know more about my teaching activity, see here.

News 1: we will be welcoming Yu-Han Wu in the team for a masters internship starting from April 2023.
News 2: I will be joining Lénaïc Chizat's group at EPFL for a post-doc starting from the winter 2023-24.

A (very short) Curriculum Vitae

Education

2019-2023: Sorbonne University, M2 and PhD degrees
2015-2019: Ecole Polytechnique, Engineering degree (specialization in Applied Mathematics)
2013-2015: Lycée du Parc, Preparatory classes

Internships

2020: Google, Research Internship
2018-2019: Engie, Business Development Internship
2018: University of Montreal, Research Internship
2017-2018: Société Générale, Software Engineering Internship
2015-2016: EPIDE, Teaching and Mentoring Internship

Some cool software projects I worked on

Building custom Jupyter widgets

Jupyter widgets enable interactive data visualization in the Jupyter notebooks.

Computing highly-uniform point sets

LatNet Builder is a state-of-the-art C++ software tool for constructing highly-uniform point sets for quasi-Monte Carlo methods.

Question Answering grounded in Knowledge Graphs

Knowledge Graphs are a common format to store real-world information at scale, and querying them to answer factual questions is useful for voice assistants.