Research map / Learning theory and algorithms / Matrix and tensor methods
Probabilistic Linear Solvers for Machine Learning
Jonathan Wenger, Philipp Hennig
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 |
|---|---|---|---|
| add Rayleigh quotient GP regression calibration vs none (ablation) | probabilistic linear solver · Airline delay, Matern32 kernel, n=10^2 | mean calibration statistic w: -5.99 -> -0.24 | Improved |
| add Rayleigh quotient GP regression calibration vs none (ablation) | probabilistic linear solver · Airline delay, RBF kernel, n=10^4 | mean calibration statistic w: 0.14 -> 21.32 | Hurt |
| set phi to eps^2 vs none (ablation) | probabilistic linear solver · Airline delay, Matern32 kernel, n=10^3 | mean calibration statistic w: -1.93 -> 4.26 | Improved |
| set phi to eps^2 vs Rayleigh regression (ablation) | probabilistic linear solver · Airline delay, Matern52 kernel, n=10^4 | mean calibration statistic w: 10.81 -> 0.80 | Improved |
| set phi to eps^2 vs average remaining spectrum (ablation) | probabilistic linear solver · Airline delay, RBF kernel, n=10^4 | mean calibration statistic w: 2.92 -> 2.92 | No clear effect |
| set phi to eps^2 vs average remaining spectrum (ablation) | probabilistic linear solver · Airline delay, Matern52 kernel, n=10^3 | mean calibration statistic w: -0.80 -> -0.81 | No clear effect |
Closest papers by meaning
- Affine Tracing: A New Paradigm for Probabilistic Linear Solvers (NeurIPS 2026 · similarity 0.87)
- Learning to Relax: Setting Solver Parameters Across a Sequence of Linear System Instances (ICLR 2024 · similarity 0.87)
- Learning to Discover Iterative Spectral Algorithms (NeurIPS 2026 · similarity 0.86)
- Gradients of Functions of Large Matrices (NeurIPS 2024 · similarity 0.86)
- Free Decompression with Algebraic Spectral Curves (NeurIPS 2026 · similarity 0.86)
- Improving Linear System Solvers for Hyperparameter Optimisation in Iterative Gaussian Processes (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.
