activity
20182023
most citedStochastic Distributed Learning with Gradient Quantization and Variance Reduction

81 citations · 132 across the 18 of their papers we have counts for

collaborators
Showing cs.LGShow all

16 papers · 1 filter

cs.LG2023★ 3 cited

Prodigy: An Expeditiously Adaptive Parameter-Free Learner

Konstantin Mishchenko, Aaron Defazio

We consider the problem of estimating the learning rate in adaptive methods, such as AdaGrad and Adam. We propose Prodigy, an algorithm that provably estimates the distance to the…

cs.LG2023★ 1 cited

Partially Personalized Federated Learning: Breaking the Curse of Data Heterogeneity

Konstantin Mishchenko, Rustem Islamov, Eduard Gorbunov +1

We present a partially personalized formulation of Federated Learning (FL) that strikes a balance between the flexibility of personalization and cooperativeness of global training.…

cs.LG2023★ 2 cited

DoWG Unleashed: An Efficient Universal Parameter-Free Gradient Descent Method

Ahmed Khaled, Konstantin Mishchenko, Chi Jin

This paper proposes a new easy-to-implement parameter-free gradient-based optimizer: DoWG (Distance over Weighted Gradients). We prove that DoWG is efficient -- matching the conver…

cs.LG2023

Two Losses Are Better Than One: Faster Optimization Using a Cheaper Proxy

Blake Woodworth, Konstantin Mishchenko, Francis Bach

We present an algorithm for minimizing an objective with hard-to-compute gradients by using a related, easier-to-access function as a proxy. Our algorithm is based on approximate p…

cs.LG2023★ 5 cited

Learning-Rate-Free Learning by D-Adaptation

Aaron Defazio, Konstantin Mishchenko

D-Adaptation is an approach to automatically setting the learning rate which asymptotically achieves the optimal rate of convergence for minimizing convex Lipschitz functions, with…

cs.LG2022★ 3 cited

Adaptive Learning Rates for Faster Stochastic Gradient Methods

Samuel Horváth, Konstantin Mishchenko, Peter Richtárik

In this work, we propose new adaptive step size strategies that improve several stochastic gradient methods. Our first method (StoPS) is based on the classical Polyak step size (Po…