Louis THIRY
Scholar

Louis THIRY

Google Scholar ID: CrfXOCMAAAAJ
Associate Professor, Sorbonne Université (prev. Ph.D@ENS Paris, Postdoc@INRIA, Researcher@HCompany)
machine learningcomputer visiongeophysical fluid dynamicsdensity functional theoryagentic AI
Citations & Impact
All-time
Citations
906
 
H-index
9
 
i10-index
9
 
Publications
20
 
Co-authors
0
 
Publications
20 items
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
  • Published in top venues including Science, Journal of Advances in Earth Modeling Systems, Geoscientific Model Development, Physical Review Materials, J. Chem. Phys., ICLR, NeurIPS, and ICCC.
  • Developed and open-sourced multiple PyTorch implementations: multi-layer quasi-geostrophic solvers, neural style transfer (STROTSS), 3D solid-harmonic scattering transforms, Angular Fourier Series descriptors, etc.
  • Co-developed Kymatio, an open-source Python package for scattering transforms, published in Journal of Machine Learning Research (2020).
  • Presented research at the Stochastic Transport in Upper Ocean Dynamics Workshop, Imperial College London (2023).
  • Delivered a guest talk on AI and Art at the 'P versus NP' event at Futurium Museum, Berlin, featuring a live demo of AI-generated portraits drawn by a robotic arm.
Research Experience
  • 2024–2025: Researcher on LLM/VLM-based autonomous software agents at H Company, Paris.
  • 2024: Postdoc on hybrid physics-learning ocean modeling with the ANGE team at INRIA Paris.
  • 2021–2023: Postdoc on stochastic modeling and data assimilation for oceanic flows with the ODYSSEY team at INRIA Rennes.
  • 2022: Taught Statistics and Probability to first-year computer science students at ENS Rennes.
  • Led practical sessions (TPs) on Fundamentals of Deep Learning with accompanying slides.
Co-authors
0 total
Co-authors: 0 (list not available)