Samuel Horváth
Assistant Professor of Machine Learning
Research interests
Professor Horváth’s research interests lie at the intersection of mathematics, computer science, machine learning, optimization, and statistics, with a particular focus on federated learning. Email
Prior to joining MBZUAI, Professor Horváth completed his M.Sc. (2018) and Ph.D. (2022) in statistics at King Abdullah University of Science and Technology (KAUST) in Kingdom of Saudi Arabia. He has a relatively rich industrial experience obtained via research internships, including Amazon Scalable Machine Learning, Germany (2019), Samsung AI Centre, United Kingdom (2020), and Facebook AI Research, Canada (2021).
- Ph.D. in statistics, King Abdullah University of Science and Technology (KAUST), Kingdom of Saudi Arabia
- Master’s in statistics, King Abdullah University of Science and Technology (KAUST), Kingdom of Saudi Arabia
- Bachelor (Summa Cum Laude) in mathematics of economics and finance from Comenius University, Slovakia
-
- Dean’s list, Academic Excellence Award, KAUST 2022
- Al Kindi Statistics Research Student Award, KAUST 2021
- Best Poster Award at the Data Science Summer School
- Best Paper Award at the NeurIPS 2020 Federated Learning Workshop
- Best Reviewer Award at NeurIPS 2020
- Top Reviewer Award at NeurIPS 2019
- “Fedshuffle: Recipes for better use of local work in federated learning”, S Horváth, M Sanjabi, L Xiao, P Richtárik, M Rabbat"
- “Adaptivity of Stochastic Gradient Methods for Nonconvex Optimization”, S Horváth, L Lei, P Richtárik, MI Jordan,SIAM Journal on Mathematics of Data Science (SIMODS)
- “FjORD: Fair and Accurate Federated Learning under heterogeneous targets with Ordered Dropout”, S Horváth, S Laskaridis, M Almeida, I Leontiadis, SI Venieris, ND Lane, 35th Conference on Neural Information Processing Systems (NeurIPS 2021)
- “A Better Alternative to Error Feedback for Communication-Efficient Distributed Learning”, S Horváth, P Richtárik, International Conference on Learning Representations (ICLR 2021)
Related
Orchestrated efficiency: A new technique to increase model efficiency during training
Presented at ICML 2024, Horváth's new study proposes a framework to efficiently train machine learning models for…
Read more
Frontiers of federation at the AI Quorum
Leading scientists from across the world met earlier this month in Abu Dhabi for MBZUAI’s Second Workshop…
Read more
AI-empowered smart grids: accelerating the energy transition
Researchers from the Mohamed bin Zayed University for Artificial Intelligence (MBZUAI) are using a novel combination of…
Read moreContact faculty affairs
Interested in working with
our renowned faculty?
Fill out the below form and we will get back to you.