A three-phase alternating-update method for asymmetric tensor PCA achieves d to the power of k-minus-2 sample complexity with d-squared memory and improves when signal vectors align.
Scaling law for stochas- tic gradient descent in quadratically parameterized linear regression.arXiv preprint arXiv:2502.09106
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Mild Over-Parameterization Benefits Asymmetric Tensor PCA
A three-phase alternating-update method for asymmetric tensor PCA achieves d to the power of k-minus-2 sample complexity with d-squared memory and improves when signal vectors align.