Igor Sokolov
5 papers · 2021–2025 · 3 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓
🌍
Conference Polyglot · Bronze
🌉
Interdisciplinary Bridge · Bronze
🗺️
Taxonomy Completionist
(13)
🔥
Unstoppable · Bronze
Venues
Top co-authors
Research topics
Machine Learning
(5)
Optimization & Theory · Machine Learning
(5)
Mathematics & Optimization
(4)
Optimization
(4)
Deep Learning
(3)
Optimization & Theory · Deep Learning
(3)
Artificial Intelligence
(1)
Computer Science
(1)
Learning Paradigms · Artificial Intelligence
(1)
Learning Paradigms · Machine Learning
(1)
Systems
(1)
Keywords
distributed training
(3)
error feedback
(3)
non-convex optimization
(3)
stochastic gradient descent
(3)
stochastic optimization
(2)
convergence analysis
(2)
gradient compression
(2)
communication compression
(2)
variance reduction
(2)
convergence guarantee
(1)
convergence rate
(1)
random reshuffling
(1)
without-replacement sampling
(1)
stochastic method
(1)
biased estimator
(1)
federated learning
(1)
lazy aggregation
(1)
stochastic gradient
(1)
communication complexity
(1)
distributed learning
(1)
Papers
EF21 with Bells & Whistles: Six Algorithmic Extensions of Modern Error Feedback
JMLR 2025
Don't Compress Gradients in Random Reshuffling: Compress Gradient Differences
NeurIPS 2024
A Guide Through the Zoo of Biased SGD
NeurIPS 2023
3PC: Three Point Compressors for Communication-Efficient Distributed Training and a Better Theory for Lazy Aggregation
ICML 2022
EF21: A New, Simpler, Theoretically Better, and Practically Faster Error Feedback
NeurIPS 2021