CSMC integrates column subset selection with low-rank matrix completion to reduce computation for asymmetric incomplete matrices while claiming competitive accuracy on synthetic and real tasks.
Implicit regularization in nonconvex statistical estimation: Gradient descent converges linearly for phase retrieval, matrix completion and blind deconvolution
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Randomized Approach to Matrix Completion: Applications in Recommendation Systems and Image Inpainting
CSMC integrates column subset selection with low-rank matrix completion to reduce computation for asymmetric incomplete matrices while claiming competitive accuracy on synthetic and real tasks.