Research map / Learning theory and algorithms / Matrix and tensor methods
Solving Most Systems of Random Quadratic Equations
Gang Wang, Georgios Giannakis, Yousef Saad, Jie Chen
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 reweighted amplitude flow instead of the compared iterative solvers | Phase retrieval on the Galaxy image with CDP measurements · Galaxy image, CDP with K=4 masks | relative error: 1.0347; 1.0715\times 10^{-3}; 1.6870; 1.4211 | Improved |
| Use weighted maximal correlation initialization rather than the compared state-of-the-art initializers | RAF initialization · Real-valued Gaussian model, information-theoretic benchmark | relative initialization error: about 5% empirical advantage | Improved |
| Apply RAF at the information-theoretic measurement limit | RAF · Noiseless real Gaussian phase retrieval, n=2,000, m=3,999 | amplitude-based loss L(z^T): smaller than 10^{-25} in all 200 independent realizations | Improved |
| Use weighted maximal correlation initialization | RAF · Noiseless real Gaussian phase retrieval | relative initialization error | Improved |
| Use adaptively reweighted gradients that retain contributions from almost all samples | RAF · Noiseless random quadratic equations | recovery accuracy | Improved |
Closest papers by meaning
- Solving Random Systems of Quadratic Equations via Truncated Generalized Gradient Flow (NeurIPS 2016 · similarity 0.88)
- Fast Compressive Phase Retrieval under Bounded Noise (AAAI 2017 · similarity 0.85)
- Conic Optimization for Quadratic Regression Under Sparse Noise (JMLR 2020 · similarity 0.85)
- Small random initialization is akin to spectral learning: Optimization and generalization guarantees for overparameterized low-rank matrix reconstruction (NeurIPS 2021 · similarity 0.84)
- Semi-Random Matrix Completion via Flow-Based Adaptive Reweighting (NeurIPS 2024 · similarity 0.83)
- The Nuclear Route: Sharp Asymptotics of ERM in Overparameterized Quadratic Networks (NeurIPS 2025 · similarity 0.82)
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.
