Picture for Anastasia Koloskova

Anastasia Koloskova

The Privacy Power of Correlated Noise in Decentralized Learning

Add code
May 02, 2024
Viaarxiv icon

Asynchronous SGD on Graphs: a Unified Framework for Asynchronous Decentralized and Federated Optimization

Add code
Nov 01, 2023
Viaarxiv icon

Shuffle SGD is Always Better than SGD: Improved Analysis of SGD with Arbitrary Data Orders

Add code
Jun 15, 2023
Figure 1 for Shuffle SGD is Always Better than SGD: Improved Analysis of SGD with Arbitrary Data Orders
Figure 2 for Shuffle SGD is Always Better than SGD: Improved Analysis of SGD with Arbitrary Data Orders
Figure 3 for Shuffle SGD is Always Better than SGD: Improved Analysis of SGD with Arbitrary Data Orders
Figure 4 for Shuffle SGD is Always Better than SGD: Improved Analysis of SGD with Arbitrary Data Orders
Viaarxiv icon

Revisiting Gradient Clipping: Stochastic bias and tight convergence guarantees

Add code
May 02, 2023
Figure 1 for Revisiting Gradient Clipping: Stochastic bias and tight convergence guarantees
Figure 2 for Revisiting Gradient Clipping: Stochastic bias and tight convergence guarantees
Figure 3 for Revisiting Gradient Clipping: Stochastic bias and tight convergence guarantees
Viaarxiv icon

Convergence of Gradient Descent with Linearly Correlated Noise and Applications to Differentially Private Learning

Add code
Feb 02, 2023
Figure 1 for Convergence of Gradient Descent with Linearly Correlated Noise and Applications to Differentially Private Learning
Figure 2 for Convergence of Gradient Descent with Linearly Correlated Noise and Applications to Differentially Private Learning
Figure 3 for Convergence of Gradient Descent with Linearly Correlated Noise and Applications to Differentially Private Learning
Figure 4 for Convergence of Gradient Descent with Linearly Correlated Noise and Applications to Differentially Private Learning
Viaarxiv icon

Decentralized Gradient Tracking with Local Steps

Add code
Jan 03, 2023
Figure 1 for Decentralized Gradient Tracking with Local Steps
Figure 2 for Decentralized Gradient Tracking with Local Steps
Figure 3 for Decentralized Gradient Tracking with Local Steps
Figure 4 for Decentralized Gradient Tracking with Local Steps
Viaarxiv icon

Sharper Convergence Guarantees for Asynchronous SGD for Distributed and Federated Learning

Add code
Jun 16, 2022
Figure 1 for Sharper Convergence Guarantees for Asynchronous SGD for Distributed and Federated Learning
Figure 2 for Sharper Convergence Guarantees for Asynchronous SGD for Distributed and Federated Learning
Viaarxiv icon

Data-heterogeneity-aware Mixing for Decentralized Learning

Add code
Apr 13, 2022
Figure 1 for Data-heterogeneity-aware Mixing for Decentralized Learning
Figure 2 for Data-heterogeneity-aware Mixing for Decentralized Learning
Figure 3 for Data-heterogeneity-aware Mixing for Decentralized Learning
Figure 4 for Data-heterogeneity-aware Mixing for Decentralized Learning
Viaarxiv icon

An Improved Analysis of Gradient Tracking for Decentralized Machine Learning

Add code
Feb 08, 2022
Figure 1 for An Improved Analysis of Gradient Tracking for Decentralized Machine Learning
Figure 2 for An Improved Analysis of Gradient Tracking for Decentralized Machine Learning
Figure 3 for An Improved Analysis of Gradient Tracking for Decentralized Machine Learning
Figure 4 for An Improved Analysis of Gradient Tracking for Decentralized Machine Learning
Viaarxiv icon

RelaySum for Decentralized Deep Learning on Heterogeneous Data

Add code
Oct 08, 2021
Figure 1 for RelaySum for Decentralized Deep Learning on Heterogeneous Data
Figure 2 for RelaySum for Decentralized Deep Learning on Heterogeneous Data
Figure 3 for RelaySum for Decentralized Deep Learning on Heterogeneous Data
Figure 4 for RelaySum for Decentralized Deep Learning on Heterogeneous Data
Viaarxiv icon