Research map / Learning theory and algorithms / Matrix and tensor methods
MOSES: A Streaming Algorithm for Linear Dimensionality Reduction
Armin Eftekhari, Raphael Hauser, Andreas Grammenos
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 MOSES instead of GROUSE, FD and PM | streaming PCA · Mote voltage, humidity, light, temperature | subspace error over time | Improved |
| use MOSES with block size 2r instead of PM with block size 2n | streaming PCA · synthetic, n=200, r=10 | error vs time | No clear effect |
| increase block size b (ablation) | MOSES · synthetic, n=1200, r=15, T=2000 | error | No clear effect |
| increase ambient dimension n (ablation) | MOSES · synthetic, T=2000, r=15 | error | Hurt |
| increase rank r (ablation) | MOSES · synthetic, n=1200, T=2000 | error | Improved |
| use MOSES instead of GROUSE and PM for run time at large n and rank | streaming PCA · synthetic, alpha=1, T=10000 | run-time | Improved |
| use MOSES instead of FD for low rank | streaming PCA · synthetic, alpha=1, T=10000, r in {1,10} | run-time | Improved |
| use MOSES instead of GROUSE for memory | streaming PCA · Mote voltage | memory (Kb): 2896.87 -> 189.36 | Improved |
Closest papers by meaning
- Robust Subspace Approximation in a Stream (NeurIPS 2018 · similarity 0.88)
- Streaming Principal Component Analysis From Incomplete Data (JMLR 2019 · similarity 0.86)
- Online Directional Regression for Streaming Sufficient Dimension Reduction (NeurIPS 2026 · similarity 0.86)
- High-Dimensional Geometric Streaming for Nearly Low Rank Data (ICML 2024 · similarity 0.86)
- Low Precision Streaming PCA (NeurIPS 2025 · similarity 0.85)
- Oja's Algorithm for Streaming Sparse PCA (NeurIPS 2024 · 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.
