Joris Paret

Joris Paret

Postdoctoral Machine Learning Scientist

ITER Organization

About me


My main interests revolve around physics, simulation and computer science. I like to find innovative solutions to complex problems by combining my scientific knowledge with my technical skills in programming and machine learning.

Contact


Please do not hesitate to reach out with any questions or feedback you may have about my research or my personal projects.

Interests
  • Computational Physics
  • Condensed Matter Physics
  • Computer Science
  • Machine Learning
  • Game development
Education
  • PhD. in Computational Physics, 2021

    University of Montpellier

  • MSc. in Computational Physics, 2018

    University of Montpellier

  • BSc. in Theoretical Physics, 2016

    University of Montpellier

Experience

 
 
 
 
 
ITER Organization
Postdoctoral Machine Learning Engineer/Scientist
Feb. 2023 – Present Saint-Paul-lez-Durance, France

Exploration of the application of machine learning techniques for improving the verification and calibration of finite-element models when compared to operational measurements. Development of machine learning approaches for anomaly detection applied to machine monitoring (as part of the Tokamak Systems Monitor software suite).

Machine learning | Python | FEA | Ansys | Mechanical engineering | Nuclear fusion

 
 
 
 
 
Laboratoire Charles Coulomb, CNRS & Université de Montpellier
PhD Fellow
Oct. 2018 – Nov. 2021 Montpellier, France

PhD. thesis: Hidden order in disordered materials ( PDF)

Study of the emergence of local order in disordered materials (supercooled liquids, glasses) using information theory and various machine learning methods such as clustering and dimensionality reduction. Two hundred hours of teaching in programming and physics.

Computational physics | Machine learning | HPC | Python | C++ | Fortran | Teaching

 
 
 
 
 
Department of Physics, Université de Montréal
Research Assistant
Feb. 2018 – Aug. 2018 Montréal, Canada

Study of the phonon replica in the electronic structure of a FeSe monolayer on top of a SrTiO$_3$ substrate using Density Functional Theory and ab initio simulations. Courses on parallel computing (MPI, OpenMP, CUDA).
– Financed by the RQMP international internship grant program.

Computational physics | Parallel computing | Python

 
 
 
 
 
Laboratoire Charles Coulomb, CNRS & Université de Montpellier
Research Assistant
May. 2017 – Jul. 2017 Montpellier, France

Raman scattering and reflectometry of graphene samples on oxidised silicon with a thickness gradient. Development of a LabVIEW app for the automation of experimental measures.

Experimental physics | Python | LabVIEW

 
 
 
 
 
Laboratoire Charles Coulomb, CNRS & Université de Montpellier
Research Assistant
May. 2016 – Jul. 2016 Montpellier, France

Creation of heterostructures by mechanical exfoliation, transfer and stacking of 2D crystals. Raman spectroscopy and white-light reflectometry. Redaction of a user manual for an optical miscroscope for Master students.

Experimental physics | Python

 
 
 
 
 
Laboratoire Charles Coulomb, CNRS & Université de Montpellier
Research Assistant
Jun. 2015 – Jun. 2015 Montpellier, France

Numerical models and simulations of opinion dynamics on small-world networks.

Computational physics | Python | Network theory

Publications

(2022). Dimensionality reduction of local structure in glassy binary mixtures. In The Journal of Chemical Physics.

Cite Open access arXiv DOI PDF Dataset

(2021). partycls: A Python package for structural clustering. In The Journal Of Open Source Software.

Cite Open access DOI PDF GitHub

(2020). Assessing the structural heterogeneity of supercooled liquids through community inference. In The Journal of Chemical Physics.

Cite Closed access arXiv DOI PDF Dataset