Local Rademacher Complexity: sharper risk bounds with and without unlabeled samples

Neural Networks : the Official Journal of the International Neural Network Society
Luca OnetoDavide Anguita

Abstract

We derive in this paper a new Local Rademacher Complexity risk bound on the generalization ability of a model, which is able to take advantage of the availability of unlabeled samples. Moreover, this new bound improves state-of-the-art results even when no unlabeled samples are available.

References

Apr 9, 2004·Neural Computation·Lorenzo RosascoAlessandro Verri
Apr 17, 2013·Neural Networks : the Official Journal of the International Neural Network Society·Luca OnetoSandro Ridella

❮ Previous
Next ❯

Citations

Jul 31, 2016·Neural Networks : the Official Journal of the International Neural Network Society·Luca OnetoSandro Ridella
Jul 11, 2018·IEEE Transactions on Neural Networks and Learning Systems·Luca OnetoDavide Anguita

❮ Previous
Next ❯

Related Concepts

Related Feeds

Aphasia

Aphasia affects the ability to process language, including formulation and comprehension of language and speech, as well as the ability to read or write. Here is the latest research on aphasia.

Related Papers

The Nebraska Medical Journal
H A Ritter
Neural Networks : the Official Journal of the International Neural Network Society
Luca OnetoSandro Ridella
Health Data Management
Greg Gillespie
Physical Review. D
T E Clark ter Veldhuis WT
© 2022 Meta ULC. All rights reserved