Mher Safaryan - RankMe

35 papers A* 8A 2Journal 25

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	DASH: Faster Shampoo via Batched Block Preconditioning and Efficient Inverse-Root Solvers. CoRR Ionut-Vlad Modoranu, Philip Zmushko, Erik Schultheis, Mher Safaryan, Dan Alistarh
2026	J	`jnl`	LoRDO: Distributed Low-Rank Optimization with Infrequent Communication. CoRR Andrej Jovanovic, Alex Iacob, Mher Safaryan, Ionut-Vlad Modoranu, Lorenzo Sani, William F. Shen, Xinchi Qiu, Dan Alistarh, Nicholas D. Lane
2026	J	`jnl`	Towards Robust Scaling Laws for Optimizers. CoRR Alexandra Volkova, Mher Safaryan, Christoph H. Lampert, Dan Alistarh
2025	J	`jnl`	CAGE: Curvature-Aware Gradient Estimation For Accurate Quantization-Aware Training. CoRR Soroush Tabesh, Mher Safaryan, Dan Alistarh
2025	J	`jnl`	DES-LOC: Desynced Low Communication Adaptive Optimizers for Training Foundation Models. CoRR Alex Iacob, Lorenzo Sani, Mher Safaryan, Paris Giampouras, Samuel Horváth, Andrej Jovanovic, Meghdad Kurmanji, Preslav Aleksandrov, William F. Shen, Xinchi Qiu, Nicholas D. Lane
2025	J	`jnl`	GradSkip: Communication-Accelerated Local Gradient Methods with Better Computational Complexity. Trans. Mach. Learn. Res. Arto Maranjyan, Mher Safaryan, Peter Richtárik
2025	A*	`conf`	LDAdam: Adaptive Optimization from Low-Dimensional Gradient Statistics. ICLR Thomas Robert, Mher Safaryan, Ionut-Vlad Modoranu, Dan Alistarh
2025	J	`jnl`	MT-DAO: Multi-Timescale Distributed Adaptive Optimizers with Local Updates. CoRR Alex Iacob, Andrej Jovanovic, Mher Safaryan, Meghdad Kurmanji, Lorenzo Sani, Samuel Horváth, William F. Shen, Xinchi Qiu, Nicholas D. Lane
2025	J	`jnl`	SVD-Free Low-Rank Adaptive Gradient Optimization for Large Language Models. CoRR Ionut-Vlad Modoranu, Mher Safaryan, Erik Schultheis, Dan Alistarh
2025	J	`jnl`	Unified Scaling Laws for Compressed Representations. CoRR Andrei Panferov, Alexandra Volkova, Ionut-Vlad Modoranu, Vage Egiazarian, Mher Safaryan, Dan Alistarh
2024	A	`conf`	AsGrad: A Sharp Unified Analysis of Asynchronous-SGD Algorithms. AISTATS Rustem Islamov, Mher Safaryan, Dan Alistarh
2024	J	`jnl`	LDAdam: Adaptive Optimization from Low-Dimensional Gradient Statistics. CoRR Thomas Robert, Mher Safaryan, Ionut-Vlad Modoranu, Dan Alistarh
2024	A*	`conf`	MicroAdam: Accurate Adaptive Optimization with Low Space Overhead and Provable Convergence. NeurIPS Ionut-Vlad Modoranu, Mher Safaryan, Grigory Malinovsky, Eldar Kurtic, Thomas Robert, Peter Richtárik, Dan Alistarh
2024	J	`jnl`	MicroAdam: Accurate Adaptive Optimization with Low Space Overhead and Provable Convergence. CoRR Ionut-Vlad Modoranu, Mher Safaryan, Grigory Malinovsky, Eldar Kurtic, Thomas Robert, Peter Richtárik, Dan Alistarh
2024	A*	`conf`	The Iterative Optimal Brain Surgeon: Faster Sparse Recovery by Leveraging Second-Order Information. NeurIPS Diyuan Wu, Ionut-Vlad Modoranu, Mher Safaryan, Denis Kuznedelev, Dan Alistarh
2024	J	`jnl`	The Iterative Optimal Brain Surgeon: Faster Sparse Recovery by Leveraging Second-Order Information. CoRR Diyuan Wu, Ionut-Vlad Modoranu, Mher Safaryan, Denis Kuznedelev, Dan Alistarh
2023	J	`jnl`	AsGrad: A Sharp Unified Analysis of Asynchronous-SGD Algorithms. CoRR Rustem Islamov, Mher Safaryan, Dan Alistarh
2023	J	`jnl`	Distributed Newton-Type Methods with Communication Compression and Bernoulli Aggregation. Trans. Mach. Learn. Res. Rustem Islamov, Xun Qian, Slavomír Hanzely, Mher Safaryan, Peter Richtárik
2023	A*	`conf`	Knowledge Distillation Performs Partial Variance Reduction. NeurIPS Mher Safaryan, Alexandra Peste, Dan Alistarh
2023	J	`jnl`	Knowledge Distillation Performs Partial Variance Reduction. CoRR Mher Safaryan, Alexandra Peste, Dan Alistarh
2023	J	`jnl`	On Biased Compression for Distributed Learning. J. Mach. Learn. Res. Aleksandr Beznosikov, Samuel Horváth, Peter Richtárik, Mher Safaryan
2022	A	`conf`	Basis Matters: Better Communication-Efficient Second Order Methods for Federated Learning. AISTATS Xun Qian, Rustem Islamov, Mher Safaryan, Peter Richtárik
2022	J	`jnl`	Distributed Newton-Type Methods with Communication Compression and Bernoulli Aggregation. CoRR Rustem Islamov, Xun Qian, Slavomír Hanzely, Mher Safaryan, Peter Richtárik
2022	A*	`conf`	FedNL: Making Newton-Type Methods Applicable to Federated Learning. ICML Mher Safaryan, Rustem Islamov, Xun Qian, Peter Richtárik
2022	J	`jnl`	GradSkip: Communication-Accelerated Local Gradient Methods with Better Computational Complexity. CoRR Artavazd Maranjyan, Mher Safaryan, Peter Richtárik
2022	A*	`conf`	Theoretically Better and Numerically Faster Distributed Optimization with Smoothness-Aware Quantization Techniques. NeurIPS Bokun Wang, Mher Safaryan, Peter Richtárik
2021	J	`jnl`	Basis Matters: Better Communication-Efficient Second Order Methods for Federated Learning. CoRR Xun Qian, Rustem Islamov, Mher Safaryan, Peter Richtárik
2021	J	`jnl`	FedNL: Making Newton-Type Methods Applicable to Federated Learning. CoRR Mher Safaryan, Rustem Islamov, Xun Qian, Peter Richtárik
2021	A*	`conf`	Smoothness Matrices Beat Smoothness Constants: Better Communication Compression Techniques for Distributed Optimization. NeurIPS Mher Safaryan, Filip Hanzely, Peter Richtárik
2021	J	`jnl`	Smoothness Matrices Beat Smoothness Constants: Better Communication Compression Techniques for Distributed Optimization. CoRR Mher Safaryan, Filip Hanzely, Peter Richtárik
2021	J	`jnl`	Smoothness-Aware Quantization Techniques. CoRR Bokun Wang, Mher Safaryan, Peter Richtárik
2021	A*	`conf`	Stochastic Sign Descent Methods: New Algorithms and Better Theory. ICML Mher Safaryan, Peter Richtárik
2020	J	`jnl`	On Biased Compression for Distributed Learning. CoRR Aleksandr Beznosikov, Samuel Horváth, Peter Richtárik, Mher Safaryan
2020	J	`jnl`	Optimal Gradient Compression for Distributed and Federated Learning. CoRR Alyazeed Albasyoni, Mher Safaryan, Laurent Condat, Peter Richtárik
2020	J	`jnl`	Uncertainty Principle for Communication Compression in Distributed and Federated Learning and the Search for an Optimal Compressor. CoRR Mher Safaryan, Egor Shulgin, Peter Richtárik