Research map / Learning theory and algorithms / Kernels and regression theory
Optimal Convergence Rates for Distributed Nystroem Approximation
Jian Li, Yong Liu, Weiping Wang
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 |
|---|---|---|---|
| use Nystroem centers instead of random features with the same M | kernel ridge regression (M=500) · usps, pendigits, mnist classification | test error rate | Improved |
| use DNystroem instead of DKRR and DC-NY when the number of partitions is large | distributed kernel ridge regression · usps (m>=20), pendigits (m>=10), letter (m>=37), mnist (m>=72) | test error rate: m >= 20 for usps, m >= 10 for pendigits, m >= 37 for letter, m >= 72 for mnist | Improved |
| increase the number of partitions (ablation) | DKRR, DC-NY, DC-RF, DNystroem · usps, pendigits, letter, mnist | test error rate | Hurt |
| increase the number of Nystroem centers or random features M (ablation) | Nystroem and random feature methods · usps, pendigits, letter, mnist | test error rate | Improved |
Closest papers by meaning
- Distributed Nyström Kernel Learning with Communications (ICML 2021 · similarity 0.93)
- Distributed Kernel Ridge Regression with Communications (JMLR 2020 · similarity 0.93)
- Optimal Rates of Distributed Regression with Imperfect Kernels (JMLR 2021 · similarity 0.91)
- Effective Distributed Learning with Random Features: Improved Bounds and Algorithms (ICLR 2021 · similarity 0.90)
- Distributed Semi-supervised Learning with Kernel Ridge Regression (JMLR 2017 · similarity 0.90)
- Optimal Convergence for Distributed Learning with Stochastic Gradient Methods and Spectral Algorithms (JMLR 2020 · similarity 0.90)
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.
