English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Statistical learning theory provides bounds of the generalization gap, usingin particular the Vapnik-Chervonenkis dimension and the Rademacher complexity.An alternative approach, mainly studied in the statistical physics literature,is the study of generalization in simple synthetic-data models. Here we discussthe connections between these approaches and focus on the link between theRademacher complexity in statistical learning and the theories ofgeneralization for typical-case synthetic models from statistical physics,involving quantities known as Gardner capacity and ground state energy. We showthat in these models the Rademacher complexity is closely related to the groundstate energy computed by replica theories. Using this connection, one mayreinterpret many results of the literature as rigorous Rademacher bounds in avariety of models in the high-dimensional statistics limit. Somewhatsurprisingly, we also show that statistical learning theory providespredictions for the behavior of the ground-state energies in some full replicasymmetry breaking models.

Rademacher complexity and spin glasses: A link between the replica and statistical theories of learning