Fast and Accurate Estimation of Low-Rank Matrices from Noisy Measurements via Preconditioned Non-Convex Gradient Descent.
Jialun ZhangRichard Y. ZhangHong-Ming ChiuPublished in: AISTATS (2024)
Keyphrases
- accurate estimation
- low rank matrices
- noisy measurements
- nuclear norm
- low rank matrix
- low rank
- convex optimization
- matrix completion
- singular value decomposition
- norm minimization
- image sequences
- loss function
- image motion
- singular values
- semi supervised
- low rank approximation
- convex relaxation
- matrix factorization
- approximation methods
- missing data
- sparse matrix
- high dimensional data
- linear combination
- particle filter
- feature space
- image segmentation