A new p-TNN is defined via t-SVD that better approximates tensor average rank than nuclear norm for p<1, yielding an LRTC model with error bounds, momentum-accelerated solver, and global convergence under smoothness.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Tensor p-shrinkage nuclear norm for low-rank tensor completion
A new p-TNN is defined via t-SVD that better approximates tensor average rank than nuclear norm for p<1, yielding an LRTC model with error bounds, momentum-accelerated solver, and global convergence under smoothness.