Details
Title
Karimireddy, Sai Praneeth Reddy
Sciper ID
267005
Affiliated labs
MLO
Publications
Accelerating Gradient Boosting Machines
Adaptive balancing of gradient and update computation times using global geometry and approximate subproblems
Efficient Greedy Coordinate Descent for Composite Problems
Error Feedback Fixes SignSGD and other Gradient Compression Schemes
Learning from History for Byzantine Robust Optimization
On Matching Pursuit and Coordinate Descent
Practical Low-Rank Communication Compression in Decentralized Deep Learning
Quasi-Global Momentum: Accelerating Decentralized Deep Learning on Heterogeneous Data
SCAFFOLD Stochastic Controlled Averaging for Federated Learning
The Error-Feedback Framework: Better Rates for SGD with Delayed Gradients and Compressed Updates
Adaptive balancing of gradient and update computation times using global geometry and approximate subproblems
Efficient Greedy Coordinate Descent for Composite Problems
Error Feedback Fixes SignSGD and other Gradient Compression Schemes
Learning from History for Byzantine Robust Optimization
On Matching Pursuit and Coordinate Descent
Practical Low-Rank Communication Compression in Decentralized Deep Learning
Quasi-Global Momentum: Accelerating Decentralized Deep Learning on Heterogeneous Data
SCAFFOLD Stochastic Controlled Averaging for Federated Learning
The Error-Feedback Framework: Better Rates for SGD with Delayed Gradients and Compressed Updates
Link to search
All resources
Record appears in
Authorities > People