Tuan-Anh Nguyen Welcome to my personal page :)

I am an Applied Scientist at Tiime Software, where I work on Generative AI for the accounting domain. Prior to this, I obtained my PhD in Machine Learning at Grenoble Computer Science Laboratory, University Grenoble Alpes under the supervision of two amazing people Denis Trystram and Nguyen Kim Thang.

In the summer 2023, I was a visiting PhD student at EPFL, where I worked on Knowledge Distillation and Federated Learning. I hold Master’s degree in Random Modelling from the University Paris Diderot and a Bachelor’s Degree in Mathematics from Sorbonne University

Currently, I’m interested in multi-agent systems, optimization/machine learning and bandit problems.

Home

news

  • November 2026 - New side project: PickleStats, a free statistics page that analyzes your pickleball game from your DUPR matches. Rating trends, form vs. expected results, partners, opponents and concrete tips to improve. Everything runs in your browser.
  • March 2025 - Our paper "Harnessing Increased Client Participation with Cohort-Parallel Federated Learning" is accepted to the 5th Workshop on Machine Learning and Systems (EuroMLSys 2025). This work is done during my internship at EPFL where we explore knowledge distillation and one-shot learning to improve the client participation in large-scale federated learning.
  • Feb 2025 - My thesis manuscript is officially online, you can check it out.
  • October 2024 - I recently joined the GenAI team at Tiime Software where we will focus on building AI-powered algorithm for the accounting domain.
  • October 2024 - My PhD defense will take place on the 23rd October 2024 at Maison Jean Kuntzmann, amphitheater Jean Kuntzmann, University Grenoble-Alpes. The defense will start at 2pm.
  • July 2024 - I will give a few introductory lectures on AI at the GINSEN Summer School, organized by MIAI and Grenoble INP. The topics will span general machine learning, some basic neural network architectures and transformer.
  • May 2024 - Our research paper "Handling Delayed Feedback in Distributed Online Optimization : A Projection-Free approach" has been accepted to #ECMLPKDD2024. See you in Vilnius this September.
  • May 2024 - Our paper on "Cohort Parallel Federated Learning" is available on arxiv.
  • May 2024 - We successfully organised the 16th edition of "New challenges in scheduling theory" seminar, Aussois, France.
  • Dec 2023 - I gave a presentation on online optimization in distributed setting during the Frugal AI Workshop in Grenoble.
  • Nov 2023 - I participated in the FedMalin seminar in Nice, gave a small talk on online distributed learning.
  • Sep 2023 - Talk at Inria-EPFL workshop about knowledge distillation in Federated Learning.