Research map / Learning theory and algorithms / Matrix and tensor methods
Vector-Valued Least-Squares Regression under Output Regularity Assumptions
Luc Brogat-Motte, Alessandro Rudi, Céline Brouard, Juho Rousu, Florence d'Alché-Buc
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 |
|---|---|---|---|
| replace full-rank ridge (p=infinity) with reduced-rank estimation | IOKR · USPS digit reconstruction (n=1000) | RBF loss: 0.751 -> 0.734 | Improved |
| replace SPEN energy network with reduced-rank IOKR | structured predictor · USPS digit reconstruction (n=1000) | RBF loss: 0.801 -> 0.734 | Improved |
| reduced-rank estimation on full Bibtex/Bookmarks | IOKR · Bibtex | F1: 44.0 -> 43.8 | No clear effect |
| reduced-rank estimation on full Bookmarks | IOKR · Bookmarks | F1: 39.3 -> 39.1 | No clear effect |
| reduced-rank estimation with 2000 training points | IOKR · Bibtex (n=2000) | F1: 35.9 -> 39.7 | Improved |
| reduced-rank estimation with 2000 training points | IOKR · Bookmarks (n=2000) | F1: 22.9 -> 25.9 | Improved |
| reduced-rank estimation with 2000 training points | IOKR · Corel5k (n=2000) | F1: 13.7 -> 16.1 | Improved |
| reduced-rank estimation | IOKR · Metabolite identification | Tanimoto-Gaussian loss: 0.463 -> 0.459 | Improved |
| reduced-rank estimation | IOKR · Metabolite identification | top-1 accuracy: 29.6 -> 30.0 | Improved |
| reduced-rank estimation | IOKR · Metabolite identification | MSE: 0.781 -> 0.766 | Improved |
| reduced-rank decoding on Bookmarks reduces decoding time | IOKR · Bookmarks | decoding time (s): 3371s -> 214s | Improved |
| reduced-rank decoding on Bibtex | IOKR · Bibtex | decoding time (s): 13s -> 4s | Improved |
Closest papers by meaning
- Adaptive Reduced Rank Regression (NeurIPS 2020 · similarity 0.88)
- Low-Rank-Sparse Subspace Representation for Robust Regression (CVPR 2017 · similarity 0.87)
- Towards Optimal Sobolev Norm Rates for the Vector-Valued Regularized Least-Squares Algorithm (JMLR 2024 · similarity 0.86)
- Optimal Rates for Vector-Valued Spectral Regularization Learning Algorithms (NeurIPS 2024 · similarity 0.86)
- Generalized Tensor Decomposition for Understanding Multi-Output Regression under Combinatorial Shifts (NeurIPS 2024 · similarity 0.85)
- Solving Sparse \& High-Dimensional-Output Regression via Compression (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.
