Research map / Learning theory and algorithms / Matrix and tensor methods
Fast Stagewise Sparse Factor Regression
Kun Chen, Ruipeng Dong, Wanwan Xu, Zemin Zheng
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 stagewise learning instead of ACS for sequential pursuit | SeqACS · Simulation Model II SNR=0.5 p=400 | time (s): 12.47 -> 1.03 | Improved |
| use stagewise learning instead of ACS for sequential pursuit | SeqACS · Simulation Model II SNR=0.5 p=400 | Er(XC) x10^3: 115.37 -> 57.48 | Improved |
| use stagewise learning with RRR initialization instead of ACS | ParACS(R) · Simulation Model II SNR=0.5 p=400 | time (s): 12.83 -> 0.51 | Improved |
| use stagewise learning with RRR initialization instead of ACS | ParACS(R) · Simulation Model II SNR=0.5 p=400 | Er(XC) x10^3: 106.39 -> 61.04 | Improved |
| use stagewise learning instead of ACS | SeqACS · Yeast eQTL with marginal screening | time (s): 29.84 -> 3.26 | Improved |
| use stagewise ParSTL(R) instead of RSSVD | RSSVD · Yeast eQTL full data | time (s): 1330.67 -> 36.02 | Improved |
| use stagewise ParSTL(R) instead of SeqACS | SeqACS · Yeast eQTL full data | MSE: 0.26 -> 0.22 | Improved |
| use sparse low-rank ParSTL(R) instead of plain RRR | RRR · Yeast eQTL with marginal screening | MSE: 0.34 -> 0.21 | Improved |
| drop marginal screening, use all 3244 markers (ablation) | ParSTL(R) · Yeast eQTL | MSE: 0.21 -> 0.22 | Hurt |
| drop marginal screening, use all 3244 markers | ParSTL(R) · Yeast eQTL | time (s): 1.12 -> 36.02 | Hurt |
| use parallel pursuit instead of sequential pursuit | ACS · Simulation, sparser C* | estimation accuracy | Improved |
| use parallel instead of sequential pursuit | ACS · Simulation, less sparse C* | estimation accuracy | No clear effect |
Closest papers by meaning
- Constrained Factorization with Diagonal Scaling: Rank-Revealing Training and Pruning (NeurIPS 2026 · similarity 0.87)
- Dependent relevance determination for smooth and structured sparse regression (JMLR 2019 · similarity 0.87)
- Structured Low-Rank Matrix Factorization: Global Optimality, Algorithms, and Applications (TPAMI 2019 · similarity 0.87)
- Sparse Plus Low Rank Matrix Decomposition: A Discrete Optimization Approach (JMLR 2023 · similarity 0.87)
- Exponentially Convergent Algorithms for Supervised Matrix Factorization (NeurIPS 2023 · similarity 0.87)
- Two Sparse Matrices are Better than One: Sparsifying Neural Networks with Double Sparse Factorization (ICLR 2025 · similarity 0.86)
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.
