Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5773528 | Applied and Computational Harmonic Analysis | 2018 | 21 Pages |
Abstract
The correntropy-induced loss (C-loss) has been employed in learning algorithms to improve their robustness to non-Gaussian noise and outliers recently. Despite its success on robust learning, only little work has been done to study the generalization performance of regularized regression with the C-loss. To enrich this theme, this paper investigates a kernel-based regression algorithm with the C-loss and â1-regularizer in data dependent hypothesis spaces. The asymptotic learning rate is established for the proposed algorithm in terms of novel error decomposition and capacity-based analysis technique. The sparsity characterization of the derived predictor is studied theoretically. Empirical evaluations demonstrate its advantages over the related approaches.
Related Topics
Physical Sciences and Engineering
Mathematics
Analysis
Authors
Hong Chen, Yulong Wang,