Satyen Kale

Decision Sciences · Social Sciences

36h-index8.2kcitations121works0.32yr avg
OpenAlex↗Google Scholar↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Advanced Bandit Algorithms Research(53), Machine Learning and Algorithms(43), Stochastic Gradient Optimization Techniques(30), Optimization and Search Problems(28), Sparse and Compressive Sensing Techniques(21)

Publications121 total

Faster Rates For Federated Variational Inequalities
arXiv (Cornell University)·2026Open Access
Faster Rates For Federated Variational Inequalities
ArXiv.org·2026Open Access
Understanding Outer Optimizers in Local SGD: Learning Rates, Momentum, and Acceleration
arXiv (Cornell University)·2025Open Access
Eager Updates For Overlapped Communication and Computation in DiLoCo
arXiv (Cornell University)·2025Open Access
Streaming DiLoCo with overlapping communication: Towards a Distributed Free Lunch
arXiv (Cornell University)·2025Open Access
Almost Tight Bounds for Differentially Private Densest Subgraph
Society for Industrial and Applied Mathematics eBooks·2025· 3 cited
Stacking as Accelerated Gradient Descent
arXiv (Cornell University)·2024Open Access
Efficient Stagewise Pretraining via Progressive Subnetworks
arXiv (Cornell University)·2024Open Access
Asynchronous Local-SGD Training for Language Modeling
arXiv (Cornell University)·2024Open Access
Improved Differentially Private and Lazy Online Convex Optimization
arXiv (Cornell University)·2023Open Access
Almost Tight Bounds for Differentially Private Densest Subgraph
arXiv (Cornell University)·2023Open Access
On the Convergence of Federated Averaging with Cyclic Client Participation
arXiv (Cornell University)·2023· 7 citedOpen Access
From Gradient Flow on Population Loss to Learning with Stochastic Gradient Descent
arXiv (Cornell University)·2022Open Access
Private Matrix Approximation and Geometry of Unitary Orbits
arXiv (Cornell University)·2022Open Access
Beyond Uniform Lipschitz Condition in Differentially Private Optimization
arXiv (Cornell University)·2022· 1 citedOpen Access
On the Unreasonable Effectiveness of Federated Averaging with Heterogeneous Data
arXiv (Cornell University)·2022· 19 citedOpen Access
Self-Consistency of the Fokker-Planck Equation
arXiv (Cornell University)·2022· 5 citedOpen Access
Mixed Federated Learning: Joint Decentralized and Centralized Learning
arXiv (Cornell University)·2022· 5 citedOpen Access
Reproducibility in Optimization: Theoretical Framework and Limits
arXiv (Cornell University)·2022· 5 citedOpen Access
Pushing the Efficiency-Regret Pareto Frontier for Online Learning of Portfolios and Quantum States
arXiv (Cornell University)·2022· 3 citedOpen Access
page 1 of 7Next →

Frequent Co-authors

Elad Hazan(20), Sashank J. Reddi(10), Sanjiv Kumar(5), Sanjeev Arora(5), John Langford(4), Robert E. Schapire(4), Sai Praneeth Karimireddy(4), Mehryar Mohri(4), Sebastian U. Stich(4), Ananda Theertha Suresh(4), Amit Agarwal(3), Boaz Barak(3), Daniel Hsu(3), Martin Jaggi(3), Miroslav Dudı́k(3), Lev Reyzin(3), Alekh Agarwal(2), Felix X. Yu(2), Nikos Karampatziakis(2), Moritz Hardt(2)