| 2025 | AISTATS | DPFL: Decentralized Personalized Federated Learning. | Salma Kharrat, Marco Canini, Samuel Horvth |
| 2025 | AISTATS | Revisiting LocalSGD and SCAFFOLD: Improved Rates and Missing Analysis. | Ruichen Luo, Sebastian U. Stich, Samuel Horvth, Martin Takc |
| 2025 | ECAI | Quantize Once, Train Fast: Allreduce-Compatible Compression with Provable Guarantees. | Jihao Xin, Marco Canini, Peter Richtrik, Samuel Horvth |
| 2025 | ICLR | Methods with Local Steps and Random Reshuffling for Generally Smooth Non-Convex Federated Optimization. | Yury Demidovich, Petr Ostroukhov, Grigory Malinovsky, Samuel Horvth, Martin Takc, Peter Richtrik, Eduard Gorbunov |
| 2025 | ICLR | Methods for Convex (L0, L1)-Smooth Optimization: Clipping, Acceleration, and Adaptivity. | Eduard Gorbunov, Nazarii Tupitsa, Sayantan Choudhury, Alen Aliev, Peter Richtrik, Samuel Horvth, Martin Takc |
| 2025 | ICML | Clipping Improves Adam-Norm and AdaGrad-Norm when the Noise Is Heavy-Tailed. | Savelii Chezhegov, Yaroslav Klyukin, Andrei Semenov, Aleksandr Beznosikov, Alexander V. Gasnikov, Samuel Horvth, Martin Takc, Eduard Gorbunov |
| 2025 | ICML | Aequa: Fair Model Rewards in Collaborative Learning via Slimmable Networks. | Nurbek Tastan, Samuel Horvth, Karthik Nandakumar |
| 2025 | ICML | FRUGAL: Memory-Efficient Optimization by Reducing State Overhead for Scalable Training. | Philip Zmushko, Aleksandr Beznosikov, Martin Takc, Samuel Horvth |
| 2024 | AISTATS | Efficient Conformal Prediction under Data Heterogeneity. | Vincent Plassier, Nikita Kotelevskii, Aleksandr Rubashevskii, Fedor Noskov, Maksim Velikanov, Alexander Fishkov, Samuel Horvth, Martin Takc, Eric Moulines, Maxim Panov |
| 2024 | EMNLP | Low-Resource Machine Translation through the Lens of Personalized Federated Learning. | Viktor Moskvoretskii, Nazarii Tupitsa, Chris Biemann, Samuel Horvth, Eduard Gorbunov, Irina Nikishina |
| 2024 | ICML | High-Probability Convergence for Composite and Distributed Stochastic Minimization and Variational Inequalities with Heavy-Tailed Noise. | Eduard Gorbunov, Abdurakhmon Sadiev, Marina Danilova, Samuel Horvth, Gauthier Gidel, Pavel E. Dvurechensky, Alexander V. Gasnikov, Peter Richtrik |
| 2024 | ICML | Maestro: Uncovering Low-Rank Structures via Trainable Decomposition. | Samuel Horvth, Stefanos Laskaridis, Shashank Rajput, Hongyi Wang |
| 2024 | IJCAI | Dirichlet-based Uncertainty Quantification for Personalized Federated Learning with Improved Posterior Networks. | Nikita Kotelevskii, Samuel Horvth, Karthik Nandakumar, Martin Takc, Maxim Panov |
| 2024 | IJCAI | Redefining Contributions: Shapley-Driven Federated Learning. | Nurbek Tastan, Samar Fares, Toluwani Aremu, Samuel Horvth, Karthik Nandakumar |
| 2023 | ICLR | Variance Reduction is an Antidote to Byzantines: Better Rates, Weaker Assumptions and Communication Compression as a Cherry on the Top. | Eduard Gorbunov, Samuel Horvth, Peter Richtrik, Gauthier Gidel |
| 2023 | ICML | Convergence of Proximal Point and Extragradient-Based Methods Beyond Monotonicity: the Case of Negative Comonotonicity. | Eduard Gorbunov, Adrien B. Taylor, Samuel Horvth, Gauthier Gidel |
| 2023 | ICML | High-Probability Bounds for Stochastic Optimization and Variational Inequalities: the Case of Unbounded Variance. | Abdurakhmon Sadiev, Marina Danilova, Eduard Gorbunov, Samuel Horvth, Gauthier Gidel, Pavel E. Dvurechensky, Alexander V. Gasnikov, Peter Richtrik |
| 2022 | AISTATS | FLIX: A Simple and Communication-Efficient Alternative to Local Methods in Federated Learning. | Elnur Gasanov, Ahmed Khaled, Samuel Horvth, Peter Richtrik |
| 2021 | AISTATS | Hyperparameter Transfer Learning with Adaptive Complexity. | Samuel Horvth, Aaron Klein, Peter Richtrik, Cdric Archambeau |
| 2021 | CoNEXT | FL_PyTorch: optimization research simulator for federated learning. | Konstantin Burlachenko, Samuel Horvth, Peter Richtrik |
| 2021 | ICLR | A Better Alternative to Error Feedback for Communication-Efficient Distributed Learning. | Samuel Horvth, Peter Richtrik |
| 2020 | ALT | Don't Jump Through Hoops and Remove Those Loops: SVRG and Katyusha are Better Without the Outer Loop. | Dmitry Kovalev, Samuel Horvth, Peter Richtrik |
| 2019 | ICML | Nonconvex Variance Reduced Optimization with Arbitrary Sampling. | Samuel Horvth, Peter Richtrik |