PR
Peter Richtárik
math.OCcs.LGstat.MLcs.DCcs.NAmath.NAcs.AIcs.DScs.CRcs.CV
On Valency
published · living versionsW_a96hpqgw·v1 · currentpublished
Federated Optimization: Distributed Machine Learning for On-Device Intelligence
with Jakub Konečný, H. Brendan McMahan, Daniel Ramage
1 version
Preprints & journals
291 papers in the corpus · 2008–2026Broximal Gradient Descent: A Projection-Free Sister of Projected Gradient Descent2605.08850v2 · Peter Richt'arik, Kaja Gruntkowska, Hanmin Li2026 · 0 citationsarXiv
Optimizers for Diffusion Models: A Controlled Benchmark2609.23055v1 · Arman Bolatov, Egor Shulgin, David Li et al.2026 · 0 citationsarXiv
Super-Tuning: From Activation-Aware Pruning to Sparse Fine-Tuning2607.09287v1 · Ivan Ilin, Philip Zmushko, Peter Richt'arik2026 · 0 citationsarXiv
Convergence Analysis of the ProbAbilistic Gradient Estimator Algorithm for Weakly Convex Finite-Sum Optimization2509.00737v3 · Laurent Condat, Peter Richt'arik2025 · 0 citationsarXiv
Convergence Analysis of Muon-type Methods with Inexact LMO in the Degenerate Case2606.21581v1 · Xun Qian, Peter Richt'arik2026 · 0 citationsarXiv
SILAGE: Memory-Efficient, Full-Gradient-Free Nonconvex Optimization for Nested Finite Sums2606.15832v2 · Igor Sokolov, Laurent Condat, Peter Richt'arik2026 · 0 citationsarXiv
Stochastic Sign Descent Methods: New Algorithms and Better Theory1905.12938v6 · Mher Safaryan, Peter Richt'arik2019 · 10 citationsarXiv
TAMUNA: Doubly Accelerated Distributed Optimization under Partial Participation2302.09832v4 · Laurent Condat, Ivan Agarsk'y, Grigory Malinovsky et al.2023 · 3 citationsarXiv
Separable Approximations and Decomposition Methods for the Augmented Lagrangian1308.6774v1 · Rachael Tappenden, Peter Richtarik, Burak Buke2013 · 33 citationsarXiv
Semi-Stochastic Coordinate Descent1412.6293v1 · Jakub Konecn'y, Zheng Qu, Peter Richt'arik2014 · 91 citationsarXiv
Even Faster Accelerated Coordinate Descent Using Non-Uniform Sampling1512.09103v3 · Zeyuan Allen-Zhu, Zheng Qu, Peter Richt'arik et al.2015 · 109 citationsarXiv
Randomized Distributed Mean Estimation: Accuracy vs Communication1611.07555v1 · Jakub Konecn'y, Peter Richt'arik2016 · 82 citationsarXiv
Linearly Convergent Randomized Iterative Methods for Computing the Pseudoinverse1612.06255v3 · Robert M. Gower, Peter Richt'arik2016 · 20 citationsarXiv
Parallel Stochastic Newton Method1705.02005v3 · Mojm'ir Mutn'y, Peter Richt'arik2017 · 3 citationsarXiv
A Batch-Incremental Video Background Estimation Model using Weighted Low-Rank Approximation of Matrices1707.00281v1 · Aritra Dutta, Xin Li, Peter Richt'arik2017 · 9 citationsarXiv
Linearly convergent stochastic heavy ball method for minimizing generalization error1710.10737v2 · Nicolas Loizou, Peter Richt'arik2017 · 25 citationsarXiv
The Complexity of Primal-Dual Fixed Point Methods for Ridge Regression1801.06354v1 · Ademir Alves Riberio, Peter Richt'arik2018 · 1 citationarXiv
Accelerated Stochastic Matrix Inversion: General Theory and Speeding up BFGS Rules for Faster Second-Order Optimization1802.04079v2 · Robert M. Gower, Filip Hanzely, Peter Richt'arik et al.2018 · 15 citationsarXiv
Accelerated Gossip via Stochastic Heavy Ball Method1809.08657v1 · Nicolas Loizou, Peter Richt'arik2018 · 25 citationsarXiv
Provably Accelerated Randomized Gossip Algorithms1810.13084v1 · Nicolas Loizou, Michael Rabbat, Peter Richt'arik2018 · 19 citationsarXiv
Revisiting Randomized Gossip Algorithms: General Framework, Convergence Rates and Novel Block and Accelerated Protocols1905.08645v2 · Nicolas Loizou, Peter Richt'arik2019 · 33 citationsarXiv
A Unified Theory of SGD: Variance Reduction, Sampling, Quantization and Coordinate Descent1905.11261v1 · Eduard Gorbunov, Filip Hanzely, Peter Richt'arik2019 · 26 citationsarXiv
One Method to Rule Them All: Variance Reduction for Data, Parameters and Many New Methods1905.11266v2 · Filip Hanzely, Peter Richt'arik2019 · 10 citationsarXiv
Demystifying Pipeline Parallelism: First Theory for PipeDream2606.03498v1 · Ivan Ilin, Peter Richt'arik2026 · 0 citationsarXiv
A Unified Primal-Dual Recipe for Accelerating Three-Operator Splitting Methods2605.26985v1 · Abdurakhmon Sadiev, Laurent Condat, Peter Richt'arik2026 · 0 citationsarXiv
LOSCAR-SGD: Local SGD with Communication-Computation Overlap and Delay-Corrected Sparse Model Averaging2605.20866v1 · Yassine Maziane, Ammar Mahran, Artavazd Maranjyan et al.2026 · 0 citationsarXiv
Distance-Aware Muon: Adaptive Step Scaling for Normalized Optimization2605.18999v1 · Yury Demidovich, Abhishek Chakraborty, Grigory Malinovsky et al.2026 · 0 citationsarXiv
Ringmaster LMO: Asynchronous Linear Minimization Oracle Momentum Method2605.18174v1 · Abdurakhmon Sadiev, Artavazd Maranjyan, Ivan Ilin et al.2026 · 0 citationsarXiv
Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity2605.13434v1 · Ammar Mahran, Artavazd Maranjyan, Peter Richt'arik2026 · 0 citationsarXiv
Consensus-Based Optimization with Truncated Noise2310.16610v2 · Massimo Fornasier, Peter Richt'arik, Konstantin Riedl et al.2023 · 4 citationsEur. J. Appl. Math. 36 (2), 292-315, 2025
Rennala MVR: Improved Time Complexity for Parallel Stochastic Optimization via Momentum-Based Variance Reduction2605.08871v1 · Zhirayr Tovmasyan, Artavazd Maranjyan, Peter Richt'arik2026 · 0 citationsarXiv
Tighter Performance Theory of FedExProx2410.15368v2 · Wojciech Anyszka, Kaja Gruntkowska, Alexander Tyurin et al.2024 · 0 citationsarXiv
Broximal Alignment for Global Non-Convex Optimization2604.13483v1 · Kaja Gruntkowska, Hanmin Li, Xun Qian et al.2026 · 0 citationsarXiv
Communication-Efficient Gluon in Federated Learning2604.10689v1 · Xun Qian, Alexander Gaponov, Grigory Malinovsky et al.2026 · 0 citationsarXiv
A Nesterov-Accelerated Primal-Dual Splitting Algorithm for Convex Nonsmooth Optimization2604.09245v1 · Laurent Condat, Abdurakhmon Sadiev, Peter Richt'arik2026 · 0 citationsarXiv
Byzantine-Robust and Differentially Private Federated Optimization under Weaker Assumptions2603.23472v2 · Rustem Islamov, Grigory Malinovsky, Alexander Gaponov et al.2026 · 0 citationsarXiv
Stabilized Proximal Point Method via Trust Region Control2604.02943v1 · Hanmin Li, Kaja Gruntkowska, Peter Richt'arik2026 · 0 citationsarXiv
CompressedScaffnew: The First Theoretical Double Acceleration of Communication from Local Training and Compression in Distributed Optimization2210.13277v4 · Laurent Condat, Ivan Agarsk'y, Peter Richt'arik2022 · 0 citationsarXiv
Tight Lower Bounds and Optimal Algorithms for Stochastic Nonconvex Optimization with Heavy-Tailed Noise2512.18713v2 · Adrien Fradin, Abdurakhmon Sadiev, Laurent Condat et al.2025 · 0 citationsarXiv
Double Momentum and Error Feedback for Clipping with Fast Rates and Differential Privacy2502.11682v2 · Rustem Islamov, Samuel Horvath, Aurelien Lucchi et al.2025 · 0 citationsarXiv
Revisiting Stochastic Proximal Point Methods: Generalized Smoothness and Similarity2502.03401v2 · Zhirayr Tovmasyan, Grigory Malinovsky, Laurent Condat et al.2025 · 1 citationarXiv
BiCoLoR: Communication-Efficient Optimization with Bidirectional Compression and Local Training2601.12400v1 · Laurent Condat, Artavazd Maranjyan, Peter Richt'arik2026 · 0 citationsarXiv
MAST: Model-Agnostic Sparsified Training2311.16086v2 · Yury Demidovich, Grigory Malinovsky, Egor Shulgin et al.2023 · 0 citationsarXiv
First Provable Guarantees for Practical Private FL: Beyond Restrictive Assumptions2512.21521v1 · Egor Shulgin, Grigory Malinovsky, Sarit Khirirat et al.2025 · 0 citationsarXiv
Better LMO-based Momentum Methods with Second-Order Information2512.13227v1 · Sarit Khirirat, Abdurakhmon Sadiev, Yury Demidovich et al.2025 · 0 citationsarXiv
Beyond the Ideal: Analyzing the Inexact Muon Update2510.19933v1 · Egor Shulgin, Sultan AlRashed, Francesco Orabona et al.2025 · 0 citationsarXiv
Second-order Optimization under Heavy-Tailed Noise: Hessian Clipping and Sample Complexity Limits2510.10690v1 · Abdurakhmon Sadiev, Peter Richt'arik, Ilyas Fatkhullin2025 · 0 citationsarXiv
Drop-Muon: Update Less, Converge Faster2510.02239v1 · Kaja Gruntkowska, Yassine Maziane, Zheng Qu et al.2025 · 0 citationsarXiv
Error Feedback for Muon and Friends2510.00643v1 · Kaja Gruntkowska, Alexander Gaponov, Zhirayr Tovmasyan et al.2025 · 0 citationsarXiv
Non-Euclidean Broximal Point Method: A Blueprint for Geometry-Aware Optimization2510.00823v1 · Kaja Gruntkowska, Peter Richt'arik2025 · 0 citationsarXiv
Career total: 391 works. 291 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.