About me

My name is Avetik Karagulyan. I am a Research Scientist at CNRS/L2S. Previously, I was a PostDoctoral fellow at KAUST in the team of professor Peter Richtárik. I defended my thesis at Center of Research in Economics and Statistics (CREST), Paris under the supervision of professor Arnak Dalalyan.

In 2018, I received my MSc Mathematics, Vision, Learning (MVA) diploma at ENS Paris-Saclay with highest honors (mention "très bien"). I graduated from Yerevan State University's faculty of Mathematics and Mechanics in 2017 with excellence.

My research focuses on the study of different methods of sampling and their connections to optimization.


News

2026 Apr - I am co-organizing the 11th Journée Statistique et Informatique pour la Science des Données with Erwan Le Pennec on April 3, 2026, at IHES. Registration is free and open until March 27, 2026.


2025 Nov - Was invited to give a talk at the Department of Statistics at the University of Oxford.


2025 Sep - Tigran Galstyan joined the L2S lab as a postdoc, co-supervised by me and Arshak Minasyan.


2025 June - I will attend EUROPT 2025 in Southampton, as an invited speaker in the "Optimization for Machine Learning" stream.


2025 May - Our paper on the ELF algorithm got accepted to the Uncertainty in Artificial Intelligence Conference, which will be held in Rio de Janeiro from the 21th to the 25th of July.


2025 April - I was invited to Recent Advances in Stastical Robustness workshop held at ESSEC Rabat.


2025 March - Gave a talk on cross-device federated learning at the Séminaire d'Automatique du Plateau de Saclay.


2025 March - With Arshak Minasyan, we are looking for a postdoc, interested in working on certain topics of Statistics and Optimization. The application deadline is mid-March. The details can be found here.


2025 February - Gave a talk on federated sampling at The Paris-Saclay Signal Seminar (Scube).


2024 November - I will join Laboratoire des Signaux et Systèmes (L2S) at Centrale Supélec as a Research Scientist (Chargé de Recherche) from Centre Nationale de la Recherche Scientifique (CNRS).


2024 May - "SPAM: Stochastic Proximal Point Method with Momentum Variance Reduction for Non-convex Cross-Device Federated Learning" is on arXiv. It is a joint work with Egor Shulgin, Abdurakhmon Sadiev, and Peter Richtárik.


2024 April - Two papers accepted to ICLR!