Sayantan Choudhury

Hi! I am Sayantan Choudhury, a Postdoctoral Researcher at MBZUAI, hosted by Martin Takac and Mladen Kolar. I completed my PhD from the Department of Applied Mathematics and Statistics at Johns Hopkins University. I am very fortunate to work under the supervision of Nicolas Loizou.

My research is in the area of Optimization and Machine Learning. My work analyses Extragradient Methods for solving structured non-convex problems and Federated Learning. I am also interested in adaptive methods and methods for solving large-scale linear equations. In Fall 2023, I was a research intern at MBZUAI, UAE, where I was working with Martin Takac on designing adaptive methods.

Before coming to JHU, I completed my B.Stat and M.Stat from Indian Statistical Institute, Kolkata. During my Bachelors and Masters, I was involved in several research projects at the Indian Statistical Institute and the Centre for Science of Student Learning. After joining JHU, I completed my MSE in Applied Mathematics and Statistics with a concentration on Optimization.

Email / CV / Google Scholar / Github / LinkedIn

Education

2025 - Ph.D. in Applied Math & Stat, Johns Hopkins University
2023 - MSE in Applied Math & Stat, Johns Hopkins University
2020 - Master's in Statistics, Indian Statistical Institute, Kolkata
2018 - Bachelor's in Statistics, Indian Statistical Institute, Kolkata

Research

Multiplayer Federated Learning: Reaching Equilibrium with Less Communication
TaeHo Yoon, Sayantan Choudhury, Nicolas Loizou.
Accepted at NeurIPS, 2025.

Paper

Methods for Convex $(L_0,L_1)$-Smooth Optimization: Clipping, Acceleration, and Adaptivity
Eduard Gorbunov, Nazarii Tupitsa, Sayantan Choudhury, Alen Aliev, Peter Richtarik, Samuel Horvath, Martin Takac.
Accepted at ICLR, 2025.

Paper

Remove that Square Root: A New Efficient Scale-Invariant Version of AdaGrad
Sayantan Choudhury, Nazarii Tupitsa, Nicolas Loizou, Samuel Horvath, Martin Takac, Eduard Gorbunov.
Accepted at NeurIPS, 2024.

Paper

Communication-Efficient Gradient Descent-Accent Methods for Distributed Variational Inequalities: Unified Analysis and Local Updates
Siqi Zhang*, Sayantan Choudhury*, Sebastian Stich, Nicolas Loizou.
* Equal Contribution.
Accepted at ICLR, 2024.

Paper

Single-Call Stochastic Extragradient Methods for Structured Non-monotone Variational Inequalities: Improved Analysis under Weaker Conditions
Sayantan Choudhury, Eduard Gorbunov, Nicolas Loizou.
Accepted at NeurIPS, 2023.

Paper

Teaching

I have worked as a Teaching Assistant for the following courses:

Spring 2023 - Iterative Algorithms in Machine Learning: Theory and Applications, Johns Hopkins University.
Spring 2023 - Optimization in Data Science, Johns Hopkins University.
Fall 2022 - Large Scale Optimization for Data Science, Johns Hopkins University.
Spring 2022 - Machine Learning II, Johns Hopkins University.
Fall 2021 - Introduction to Convexity, Johns Hopkins University.
Spring 2021 - Network Analysis and Operations Research, Johns Hopkins University.

Selected Honors & Awards

2025 - Chen Family Dissertation Fellowship for Spring 2025.
2024 - Acheson J. Duncan Fund for the Advancement of Research in Statistics.
2023 - Acheson J. Duncan Fund for the Advancement of Research in Statistics.
2023 - NeurIPS 2023 Scholar Award.
2022 - MINDS Fellowship.
2019 - Award for Excellent Academic Performance in Masters First Year at Indian Statistical Institute, Kolkata.
2015 - KVPY Fellowship, 5-year scholarship for Bachelors and Masters.
2015 - Selected for INSPIRE Fellowship.

Invited Talks & Posters

2024 - Conference on Neural Information Processing Systems (NeurIPS 2024), Vancouver, Canada (Poster)
2024 - Google Research, Bengaluru, India (Talk)
2023 - Conference on Neural Information Processing Systems (NeurIPS 2023), New Orleans, USA (Poster)
2023 - SIAM Conference on Optimization (OP23), Seattle, USA (Talk)
2023 - Annual Conference on Information Sciences and Systems (CISS 2023), Baltimore, USA (Talk)

Professional Services

I have served as a reviewer for the following conferences and journals.

Conferences:

Annual Conference on Neural Information Processing Systems (NeurIPS) 2025
International Conference on Machine Learning (ICML) 2025
International Conference on Machine Learning (ICML) 2024
Annual Conference on Neural Information Processing Systems (NeurIPS) 2024

Journals:

Journal of Machine Learning Research (JMLR)
Scientific Reports
IEEE Signal Processing Letters

Thanks to Jon Barron for the website's template.