A Vector-Contraction Inequality for Rademacher Complexities

By Andreas Maurer. 2016

308 citations recorded by the source corpus.

The contraction inequality for Rademacher averages is extended to Lipschitz functions with vector-valued domains, and it is also shown that in the bounding expression the Rademacher variables can be replaced by arbitrary iid symmetric and sub-gaussian variables. Example applications are given for multi-category learning, K-means clustering and learning-to-learn.

Subfields: Artificial Intelligence, Statistical Learning Theory, Pure Mathematics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

  1. Dacheng Tao
  2. Yoshua Bengio
  3. Yuanzhi Li
  4. Tengyu Ma
  5. Ruslan Salakhutdinov
  6. Zeyuan Allen-Zhu
  7. Sergey Levine
  8. Percy Liang

In Statistical Learning Theory

  1. Yuanzhi Li
  2. Samet Oymak
  3. Zeyuan Allen-Zhu
  4. M. Soltanolkotabi
  5. Behnam Neyshabur
  6. Mingchen Li
  7. Kavosh Asadi
  8. Mehryar Mohri

In Pure Mathematics

  1. Zeyuan Allen-Zhu
  2. Dacheng Tao
  3. Zalan Fabian
  4. Corinna Cortes
  5. Zhi-Hua Zhou
  6. Yann LeCun
  7. Marc Niethammer
  8. Nati Srebro

Citations in the corpus