Research map / Optimization / Nonconvex and smooth optimization
Exponentially convergent stochastic k-PCA without variance reduction
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 Matrix Krasulina as a fully online method without a variance-reduction step | Online k-PCA · VGG and MNIST | convergence before a full data pass | Helped |
| Apply Matrix Krasulina to strictly low-rank data | Matrix Krasulina · Simulated strictly low-rank data | convergence rate | Helped |
| Increase the noise-over-signal ratio | Matrix Krasulina · Simulated effectively low-rank data | convergence rate | Hurt |
| Increase the ambient data dimension while holding intrinsic dimension fixed | Matrix Krasulina · Simulated low-rank data | convergence rate | No clear effect |
Closest papers by meaning
- Convergence of Stochastic Gradient Descent for PCA (ICML 2016 · similarity 0.90)
- Stochastic PCA with $\ell_2$ and $\ell_1$ Regularization (ICML 2018 · similarity 0.89)
- An Implicit Form of Krasulina's k-PCA Update without the Orthonormality Constraint (AAAI 2020 · similarity 0.89)
- Improved Analysis of the Accelerated Noisy Power Method with Applications to Decentralized PCA (ICML 2026 · similarity 0.88)
- An Iterative Algorithm for Differentially Private $k$-PCA with Adaptive Noise (NeurIPS 2025 · similarity 0.87)
- Efficient Sparse PCA via Block-Diagonalization (ICLR 2025 · similarity 0.87)
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.