![Photo](/dpnc/files/cache/d6706d8e9dc507be8f4785c58d2396e4_f449.jpg)
Mr Guillaume QUÉTANT
Assistant Doctorant – Groupe ATLAS (co-supervisé par Prof. Voloshynovskiy)
AEM 018
E-mail
Guillaume is a PhD student in the group of Prof. Tobias Golling at the Particle Physics Department (DPNC), jointly supervised by Prof. Slava Voloshynovskiy at the Computer Vision and Multimedia Laboratory (CVML). He is a member of the RODEM project (Robust Deep Density Models for High-Energy Physics and Solar Physics), and a member and qualified author of the ATLAS Collaboration at CERN.
Guillaume graduated from the Univ. of Geneva, Switzerland, with a BSc in physics in 2017, and an MSc with an Excellence Master Fellowship in theoretical physics in 2020. He started as a PhD student at the Univ. of Geneva in 2020, where he works on machine learning for high energy physics, especially for generative modelling and anomaly detection.
Recent work
- Turbo-Sim: A generalised autoencoder framework applied to high-energy physics (NeurIPS 2021, poster and arXiv paper)
- At ATLAS (CERN): Standalone topological clustering of energy deposits in calorimeter cells (slides)
- GANDALF: Four-momentum transcoding with attention encoder, latent space matching and diffusion decoder (poster)
- PC-JeDi: Jets generation as particle clouds with a diffusion transformer network (Journal paper and arXiv paper)
- PC-Droid: Faster and better jets generation with a diffusion transformer network (Journal paper and arXiv paper)
- TURBO: Mathematical grounds and interpretations of many deep learning training methods (Journal paper and arXiv paper)
- PIPPIN: Set to set transformation (arXiv paper)
- GalaDRIEL: Corrections to galaxy simulation (ongoing)
Publications
Complete and up-to-date list of Guillaume's publications:
- Google Scholar: https://scholar.google.com/citations?user=mDxsP5UAAAAJ.
- ORCID: https://orcid.org/0000-0002-2957-3449.
- arXiv: https://arxiv.org/search/cs?searchtype=author&query=Quétant,+G
-
TURBO: The Swiss Knife of Auto-Encoders
Guillaume Quétant QUÉTANT, Yury BELOUSOV, Vitaliy KINAKH, Slava VOLOSHYNOVSKIY
Entropy 25 (10) 1471 (2023) [arXiv] -
EPiC-ly Fast Particle Cloud Generation with Flow-Matching and Diffusion
Erik BUHMANN, Cedric EWEN, Darius A. FAROUGHY, Tobias GOLLING, Gregor KASIECZKA, Matthew LEIGH, Guillaume QUÉTANT, John Andrew RAINE, Debajyoti SENGUPTA, David SHIH
(2023) [arXiv] -
PC-Droid: Faster diffusion and improved quality for particle cloud generation
Matthew LEIGH, Debajyoti SENGUPTA, John Andrew RAINE, Guillaume QUÉTANT, Tobias GOLLING
(2023) [arXiv] -
PC-JeDi: Diffusion for Particle Cloud Generation in High Energy Physics
Matthew LEIGH, Debajyoti SENGUPTA, Guillaume QUÉTANT, John Andrew RAINE, Knut ZOCH, Tobias GOLLING
submitted to SciPost (2023) [arXiv]
-
Turbo-Sim: a generalised generative model with a physical latent space
Guillaume QUÉTANT, Mariia DROZDOVA, Vitaliy KINAKH, Tobias GOLLING, Slava VOLOSHYNOVSKIY
Proceedings of workshop in the international conference on Neural Information Processing Systems (NeurIPS) (2021) [arXiv] -
Information-theoretic stochastic contrastive conditional GAN: InfoSCC-GAN
Vitaliy KINAKH, Mariia DROZDOVA, Guillaume QUÉTANT, Tobias GOLLING, Slava VOLOSHYNOVSKIY
Proceedings of workshop in the international conference on Neural Information Processing Systems (NeurIPS) (2021) [arXiv] -
Generation of data on discontinuous manifolds via continuous stochastic non-invertible networks
Mariia DROZDOVA, Vitaliy KINAKH, Guillaume QUÉTANT, Tobias GOLLING, Slava VOLOSHYNOVSKIY
Proceedings of workshop in the international conference on Neural Information Processing Systems (NeurIPS) (2021) [arXiv]
Teaching/Tutoring
Information Theory for Data Science and Machine Learning (Univ. of Geneva, 2021–2023)