Lune

Profiles2022-2026

Sebastian U Stich

CISPA Helmholtz Center for Information Security

Also published as Sebastian Stich

19Papers
12Last author
18Unique co-authors
0First author
0Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers19

  1. DADA: Dual Averaging with Distance AdaptationAuthor 4 of 4ICLR 2026 · 4 citations
  2. Enhancing LLM Training via Spectral ClippingAuthor 3 of 3ICML 2026 · 4 citations
  3. Composite Optimization with Error Feedback: the Dual Averaging ApproachAuthor 3 of 3ICLR 2026 · 2 citations
  4. Non-Convex Federated Optimization under Cost-Aware Client SelectionAuthor 3 of 3ICLR 2026 · 1 citation
  5. Revisiting Consensus Error: A Fine-grained Analysis of Local SGD under Second-order Data HeterogeneityAuthor 4 of 5NeurIPS 2025 · 1 citation
  6. Decoupled SGDA for Games with Intermittent Strategy CommunicationAuthor 5 of 5ICML 2025
  7. Optimizing (L0, L1)-Smooth Functions by Gradient MethodsAuthor 5 of 5ICLR 2025
  8. Towards Faster Decentralized Stochastic Optimization with Communication CompressionAuthor 3 of 3ICLR 2025
  9. EControl: Fast Distributed Optimization with Compression and Error ControlAuthor 3 of 3ICLR 2024 · 19 citations
  10. Non-convex Stochastic Composite Optimization with Polyak MomentumAuthor 3 of 3ICML 2024 · 13 citations
  11. Stabilized Proximal-Point Methods for Federated OptimizationAuthor 3 of 3NeurIPS 2024 · 13 citations
  12. Spectral Preconditioning for Gradient Methods on Graded Non-convex FunctionsAuthor 2 of 3ICML 2024 · 10 citations
  13. An improved analysis of per-sample and per-update clipping in federated learningAuthor 6 of 8ICLR 2024 · 9 citations
  14. Communication-Efficient Gradient Descent-Accent Methods for Distributed Variational Inequalities: Unified Analysis and Local UpdatesAuthor 3 of 4ICLR 2024 · 9 citations
  15. On Convergence of Incremental Gradient for Non-convex Smooth FunctionsAuthor 3 of 4ICML 2024 · 6 citations
  16. Adaptive SGD with Polyak stepsize and Line-search: Robust Convergence and Variance ReductionAuthor 2 of 2NeurIPS 2023 · 40 citations
  17. On the Effectiveness of Partial Variance Reduction in Federated Learning with Heterogeneous DataAuthor 5 of 5CVPR 2023
  18. Sharper Convergence Guarantees for Asynchronous SGD for Distributed and Federated LearningAuthor 2 of 3NeurIPS 2022 · 131 citations
  19. Decentralized Local Stochastic Extra-Gradient for Variational InequalitiesAuthor 5 of 6NeurIPS 2022 · 49 citations
Dusk over the sea between two cliffs drawn in fine vertical lines