Iterative regularization for learning with convex loss functions
Author(s)Lin, Junhong; Zhou, Ding-Xuan; Rosasco, Lorenzo
MetadataShow full item record
We consider the problem of supervised learning with convex loss functions and propose a new form of iterative regularization based on the subgradient method. Unlike other regularization approaches, in iterative regularization no constraint or penalization is considered, and generalization is achieved by (early) stopping an empirical iteration. We consider a nonparametric setting, in the framework of reproducing kernel Hilbert spaces, and prove consistency and finite sample bounds on the excess risk under general regularity conditions. Our study provides a new class of efficient regularized learning algorithms and gives insights on the interplay between statistics and optimization in machine learning.
DepartmentMcGovern Institute for Brain Research at MIT
Journal of Machine Learning Research
Lin, Junhong, Lorenzo Rosasaco, and Ding-Xuan Zhou. "Iterative Regularization for Learning with Convex Loss Functions." Journal of Machine Learning Research 17, 2016, pp. 1-38. © 2016 Junhong Lin, Lorenzo Rosasco and Ding-Xuan Zhou
Final published version