Sarit Khirirat

28 papers A* 2Misc 4Journal 17Unranked 4
YearRankTypeTitle / Venue / Authors
2025 J jnl
Better LMO-based Momentum Methods with Second-Order Information.
CoRR
Sarit Khirirat, Abdurakhmon Sadiev, Yury Demidovich, Peter Richtárik
2025 — conf
Collaborative Value Function Estimation Under Model Mismatch: A Federated Temporal Difference Analysis.
ECML/PKDD (6)
Ali Beikmohammadi, Sarit Khirirat, Peter Richtárik, Sindri Magnússon
2025 J jnl
Collaborative Value Function Estimation Under Model Mismatch: A Federated Temporal Difference Analysis.
CoRR
Ali Beikmohammadi, Sarit Khirirat, Peter Richtárik, Sindri Magnússon
2025 J jnl
Communication-Adaptive-Gradient Sparsification for Federated Learning With Error Compensation.
IEEE Internet Things J.
Shubham Vaishnav, Sarit Khirirat, Sindri Magnússon
2025 J jnl
First Provable Guarantees for Practical Private FL: Beyond Restrictive Assumptions.
CoRR
Egor Shulgin, Grigory Malinovsky, Sarit Khirirat, Peter Richtárik
2025 J jnl
Improved Convergence in Parameter-Agnostic Error Feedback through Momentum.
CoRR
Abdurakhmon Sadiev, Yury Demidovich, Igor Sokolov, Grigory Malinovsky, Sarit Khirirat, Peter Richtárik
2025 J jnl
On the Convergence of Federated Learning Algorithms Without Data Similarity.
IEEE Trans. Big Data
Ali Beikmohammadi, Sarit Khirirat, Sindri Magnússon
2025 J jnl
Parallel Momentum Methods Under Biased Gradient Estimations.
IEEE Trans. Control. Netw. Syst.
Ali Beikmohammadi, Sarit Khirirat, Sindri Magnússon
2025 J jnl
Smoothed Normalization for Efficient Distributed Private Optimization.
CoRR
Egor Shulgin, Sarit Khirirat, Peter Richtárik
2024 — conf
Compressed Federated Reinforcement Learning with a Generative Model.
ECML/PKDD (4)
Ali Beikmohammadi, Sarit Khirirat, Sindri Magnússon
2024 J jnl
Compressed Federated Reinforcement Learning with a Generative Model.
CoRR
Ali Beikmohammadi, Sarit Khirirat, Sindri Magnússon
2024 J jnl
Distributed Momentum Methods Under Biased Gradient Estimations.
CoRR
Ali Beikmohammadi, Sarit Khirirat, Sindri Magnússon
2024 J jnl
Error Feedback under (L
CoRR
Sarit Khirirat, Abdurakhmon Sadiev, Artem Riabinin, Eduard Gorbunov, Peter Richtárik
2024 J jnl
On the Convergence of Federated Learning Algorithms without Data Similarity.
CoRR
Ali Beikmohammadi, Sarit Khirirat, Sindri Magnússon
2023 J jnl
Clip21: Error Feedback for Gradient Clipping.
CoRR
Sarit Khirirat, Eduard Gorbunov, Samuel Horváth, Rustem Islamov, Fakhri Karray, Peter Richtárik
2023 J jnl
Improved Step-Size Schedules for Proximal Noisy Gradient Methods.
IEEE Trans. Signal Process.
Sarit Khirirat, Xiaoyu Wang, Sindri Magnússon, Mikael Johansson
2023 J jnl
Improving Performance of Private Federated Models in Medical Image Analysis.
CoRR
Xiangjian Hou, Sarit Khirirat, Mohammad Yaqub, Samuel Horváth
2022 Misc conf
Eco-Fedsplit: Federated Learning with Error-Compensated Compression.
ICASSP
Sarit Khirirat, Sindri Magnússon, Mikael Johansson
2022 — —
First-Order Algorithms for Communication Efficient Distributed Learning.
Sarit Khirirat
2022 Misc conf
Zeroth-Order Randomized Subspace Newton Methods.
ICASSP
Erik Berglund, Sarit Khirirat, Xiaoyu Wang
2021 A* conf
A Flexible Framework for Communication-Efficient Machine Learning.
AAAI
Sarit Khirirat, Sindri Magnússon, Arda Aytekin, Mikael Johansson
2021 J jnl
Compressed Gradient Methods With Hessian-Aided Error Compensation.
IEEE Trans. Signal Process.
Sarit Khirirat, Sindri Magnússon, Mikael Johansson
2021 Misc conf
Improved Step-Size Schedules for Noisy Gradient Methods.
ICASSP
Sarit Khirirat, Xiaoyu Wang, Sindri Magnússon, Mikael Johansson
2019 Misc conf
Convergence Bounds for Compressed Gradient Methods with Memory Based Error Compensation.
ICASSP
Sarit Khirirat, Sindri Magnússon, Mikael Johansson
2018 — conf
Gradient compression for communication-limited convex optimization.
CDC
Sarit Khirirat, Mikael Johansson, Dan Alistarh
2018 A* conf
The Convergence of Sparsified Gradient Methods.
NeurIPS
Dan Alistarh, Torsten Hoefler, Mikael Johansson, Nikola Konstantinov, Sarit Khirirat, Cédric Renggli
2018 J jnl
The Convergence of Sparsified Gradient Methods.
CoRR
Dan Alistarh, Torsten Hoefler, Mikael Johansson, Sarit Khirirat, Nikola Konstantinov, Cédric Renggli
2017 — conf
Mini-batch gradient descent: Faster convergence under data sparsity.
CDC
Sarit Khirirat, Hamid Reza Feyzmahdavian, Mikael Johansson