Antoine Bodin
I am a fourth year PhD student in Computer Science Department at EPFL (Swiss Federal institute of Technology of Lausanne). My research revolves around machine learning theory, statistical physics, random matrix theory, optimisation problems and artificial intelligence.
I have a Master of Science in Computer Science from EPFL and a Master of Engineering in applied Mathematics and a Minor in Financial Engineering.
Publications
Gradient flow on extensive-rank positive semi-definite matrix denoising, 2023 IEEE Information Theory Workshop (ITW)
https://arxiv.org/abs/2303.09474 / https://ieeexplore.ieee.org/document/10161669Gradient flow in the gaussian covariate model: exact solution of learning curves and multiple descent structures, preprint arxiv
https://arxiv.org/abs/2212.06757Model, sample, and epoch-wise descents: exact solution of gradient flow in the random feature model, 35th Conference on Neural Information Processing Systems (NeurIPS 2021)
https://arxiv.org/abs/2110.11805 / https://papers.nips.cc/paper/2021/hash/b4f8e5c5fb53f5ba81072451531d5460-Abstract.htmlRank-one matrix estimation: analytic time evolution of gradient descent dynamics, Conference of Learning Theory (COLT) 2021
https://arxiv.org/abs/2105.12257 / http://proceedings.mlr.press/v134/bodin21a.htmlSystems and methods for integration of disparate data feeds for unified data monitoring, US Patent (Inventor) , 2019
https://patents.google.com/patent/US11227288
Work Experience
Visiting Scholar, Harvard University (Boston MA, USA)
3 years, 6 months (Feb 2016 - July 2019)
Conducted research on the theoretical aspects of gradient descent optimization in high-dimensional spaces with Prof. Yue Lu.
Data Scientist, Credit-Suisse (Lausanne & Zurich, Switzerland)
3 years, 6 months (Feb 2016 - July 2019)
Mathematical modelling, analysis and model development in transaction surveillance for fraud detection and monitoring money laundering activity
Design and implementation of a holistic client analysis framework
Software Engineer Intern, Google (Mountain View CA, USA)
3 months (July 2015 - Sept. 2015)
Deep Learning and Natural Language Processing at Google News
Education
(2019 - Today) PhD of Computer Science - EPFL, Lausanne, Switzerland
(2014 - 2016) Master of Computer Science - EPFL, Lausanne, Switzerland
(2012 - 2014) Master of Engineering in Applied Mathematics - Centrale Paris, Paris, France
(2015 - 2016) Minor of Financial Engineering - EPFL, Lausanne, Switzerland
(2010 - 2012) Preparatory Classes MPSI/MP* - Pasteur Neuilly sur Seine, Paris, France
Dev. Skills
Python (Expert, >10years) (Tensorflow, PyTorch, Matplotlib, Django, Pandas, Numpy, Scipy, Requests, Jupyter, etc)
OCaml (>10years, occasional) (yacc, lex)
C/C++ (>10years, occasional) (OpenCL, OpenGL, GTK)
Javascript, HTML, CSS (occasional) (React, MongoDB, SQL, etc)
Contact
University: https://people.epfl.ch/antoine.bodin
GitHub: https://github.com/antoinexp
Google Scholar: https://scholar.google.com/citations?user=6PRe79MAAAAJ
Email: antoine.bodin@epfl.ch