Research map / Learning theory and algorithms / Matrix and tensor methods
The spiked matrix model with generative priors
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 a generative prior and AMP instead of an unconstrained spike prior | Spiked Wigner model · Analyzed generative-prior cases | asymptotic estimation performance | Helped |
| Use LAMP instead of canonical PCA | Spiked Wigner model with generative priors · Generative priors | noise threshold for positive correlation | Helped |
| Account for a linear generative structure with compression ratio α | Spiked Wigner model · Linear activation | noise threshold Δc: 1 -> α+1 | Helped |
| Account for a sign-activation generative structure with compression ratio α | Spiked Wigner model · Sign activation | noise threshold Δc: 1 -> 1+\frac{4\alpha}{\pi^2} | Helped |
| Use a generative prior rather than a sparse separable prior | Spiked-matrix inference · Analyzed cases in the thermodynamic limit | gap between statistical and known algorithmic performance | Helped |
| Use the empirical covariance of Fashion-MNIST spikes in the LAMP spectral method | Spiked Wigner model · Fashion-MNIST images | spike estimation performance | Helped |
Closest papers by meaning
- Spectral Phase Transition and Optimal PCA in Block-Structured Spiked Models (ICML 2024 · similarity 0.86)
- The Generative Leap: Tight Sample Complexity for Efficiently Learning Gaussian Multi-Index Models (NeurIPS 2025 · similarity 0.85)
- Efficient Algorithms for Non-gaussian Single Index Models with Generative Priors (AAAI 2024 · similarity 0.85)
- Free Decompression with Algebraic Spectral Curves (NeurIPS 2026 · similarity 0.85)
- Optimal Spectral Transitions in High-Dimensional Multi-Index Models (NeurIPS 2025 · similarity 0.85)
- Random Matrix Theory of Early-Stopped Gradient Flow: A Transient BBP Scenario (NeurIPS 2026 · 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.