Joon Kwon

INRAE Researcher
Professeur attaché — Université Paris–Saclay

Research Topics. — Machine learning, optimization, game theory, applications to life sciences.

I defended my PhD Thesis on October 18, 2016, under the supervision of Rida Laraki and Sylvain Sorin at Institut de Mathématiques de Jussieu, Université Pierre-et-Marie-Curie – Paris 6.

Distinctions

PGMO PhD Prize (2017)
Fondation mathématique Jacques Hadamard post-doc laureate (2016–2018)

Teaching

Introduction to Reinforcement Learning — M2 Université Paris-Saclay (2023–)
Online Learning, links with optimization and games — M2 Université Paris-Saclay (2024–)
Introduction to Machine Learning — CPES2 Paris Sciences et Lettres (2018–2023)

Publications

A regret minimization approach to fixed-point iterations, 2025, preprint.
On the supremum of convex lower semicontinuous functions on compact domains, 2025, preprint
Time-dependent Blackwell approachability and application to absorbing games (with Y. Wan & B. Ziliotto), preprint, 2025.
Data paper: A goat behaviour dataset combining labelled behaviours and accelerometer data for training Machine Learning detection models (with S. Mauny, N.C. Friggens, C. Duvaux-Ponter & M. Taghipoor)
Animal - Open Space, 4:100095, 2025
A pipeline with pre-processing options to detect behaviour from accelerometer data using Machine Learning tested on dairy goats (with S. Mauny, N.C. Friggens, C. Duvaux-Ponter & M. Taghipoor)
Peer Community Journal, 5(e41), 2025
Importance sampling-based gradient method for dimension reduction in Poisson log-normal model (with B. Batardière, J. Chiquet & J. Stoehr),
Electronic Journal of Statistics, 19(1):2199-2238, 2025
pyPLNmodels: A Python package to analyze multivariate high-dimensional count data (with B. Batardière, J. Chiquet & J. Stoehr)
Journal of Open Source Software, 9(104), 6969, 2024
Finite-sum optimization: Adaptivity to smoothness and loopless variance reduction (with B. Batardière), preprint, 2024.
Unifying mirror descent and dual averaging (with A. Juditsky & É. Moulines),
Mathematical Programming, 199(1-2):793-830, 2023.
Global plant extinction risk assessment informs novel biodiversity hotspots (with many co-authors), preprint, 2021.
Refined approachability algorithms and application to regret minimization with global costs,
Journal of Machine Learning Research, 22(200):1−38, 2021.
A MOM-based ensemble method for robustness, subsampling and hyperparameter tuning (with G. Lecué, M. Lerasle),
Electronic Journal of Statistics, 15(1):1202-1227, 2021.
Sparse stochastic bandits (with V. Perchet et C. Vernade),
Proc. Mach. Learn. Res. (COLT 2017), 65:1269–1270, 2017.
Online learning and Blackwell approachability with partial monitoring: optimal convergence rates (with V. Perchet),
Proc. Mach. Lean. Res. (AISTATS 2017), 54:604–613, 2017.
Gains and losses are fundamentally different in regret minimization: the sparse case (with V. Perchet),
Journal of Machine Learning Research, 17(229):1–32, 2016.
A continuous-time approach to online optimization (with P. Mertikopoulos),
Journal of Dynamics and Games, 4(2):125–148, 2017.
A universal bound on the variations of bounded convex functions,
Journal of Convex Analysis, 24(1):67–73, 2017.