An iteratively re-weighted method (IRW) is introduced for optimization problems with sparsity-inducing norms, supported by a convergence guarantee and shown to outperform alternatives on robust feature selection.
Calibrated multi-task learnin g,
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
-
An Iteratively Re-weighted Method for Problems with Sparsity-Inducing Norms
An iteratively re-weighted method (IRW) is introduced for optimization problems with sparsity-inducing norms, supported by a convergence guarantee and shown to outperform alternatives on robust feature selection.