English

Generalization Properties of Doubly Stochastic Learning Algorithms

Machine Learning 2018-03-12 v2 Machine Learning Functional Analysis Optimization and Control Statistics Theory Statistics Theory

Abstract

Doubly stochastic learning algorithms are scalable kernel methods that perform very well in practice. However, their generalization properties are not well understood and their analysis is challenging since the corresponding learning sequence may not be in the hypothesis space induced by the kernel. In this paper, we provide an in-depth theoretical analysis for different variants of doubly stochastic learning algorithms within the setting of nonparametric regression in a reproducing kernel Hilbert space and considering the square loss. Particularly, we derive convergence results on the generalization error for the studied algorithms either with or without an explicit penalty term. To the best of our knowledge, the derived results for the unregularized variants are the first of this kind, while the results for the regularized variants improve those in the literature. The novelties in our proof are a sample error bound that requires controlling the trace norm of a cumulative operator, and a refined analysis of bounding initial error.

Keywords

Cite

@article{arxiv.1707.00577,
  title  = {Generalization Properties of Doubly Stochastic Learning Algorithms},
  author = {Junhong Lin and Lorenzo Rosasco},
  journal= {arXiv preprint arXiv:1707.00577},
  year   = {2018}
}

Comments

24 pages. To appear in Journal of Complexity

R2 v1 2026-06-22T20:36:28.518Z