Research map / Optimization / Distributed and compressed
COKE: Communication-Censored Decentralized Kernel Learning
Ping Xu, Yue Wang, Xiang Chen, Zhi Tian
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add communication censoring (COKE vs DKLA) | DKLA · Twitter large | transmissions to reach MSE 2.3e-3: 1080 -> 577 | Improved |
| add communication censoring (COKE vs DKLA) | DKLA · Tom's hardware | transmissions to reach MSE 9.95e-4: 640 -> 361 | Improved |
| add communication censoring (COKE vs DKLA) | DKLA · Energy | communication cost at k=2000: 20000 -> 15752 | Improved |
| add communication censoring (COKE vs DKLA) | DKLA · Air quality | communication cost at k=2000: 20000 -> 17263 | Improved |
| add communication censoring (COKE vs DKLA) | DKLA · Twitter large | test MSE at k=50: 2.6 -> 4.2 | Hurt |
| add communication censoring (COKE vs DKLA) | DKLA · Twitter large | test MSE at k=2000: 2.5 -> 2.5 | No clear effect |
| use ADMM (DKLA) instead of diffusion CTA | CTA · Tom's hardware | test MSE at k=50: 20.16 -> 11.20 | Improved |
| use ADMM (DKLA) instead of diffusion CTA | CTA · Air quality | test MSE at k=50: 6.7 -> 2.1 | Improved |
| use COKE instead of CTA | CTA · Energy | test MSE at k=2000: 22.75 -> 21.78 | Improved |
| use COKE instead of CTA | CTA · Twitter large | communication cost at k=2000: 20000 -> 14648 | Improved |
Closest papers by meaning
- Decentralized Learning for Overparameterized Problems: A Multi-Agent Kernel Approximation Approach (ICLR 2022 · similarity 0.92)
- Decentralised Learning with Random Features and Distributed Gradient Descent (ICML 2020 · similarity 0.88)
- Hybrid Decentralized Optimization: Leveraging Both First- and Zeroth-Order Optimizers for Faster Convergence (AAAI 2025 · similarity 0.86)
- RESIST: Resilient Decentralized Learning Using Consensus Gradient Descent (NeurIPS 2026 · similarity 0.85)
- Decentralized Sporadic Federated Learning: A Unified Algorithmic Framework with Convergence Guarantees (ICLR 2025 · similarity 0.85)
- Towards Faster Decentralized Stochastic Optimization with Communication Compression (ICLR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
