Generalization of Hamiltonian algorithms

The paper proves generalization results for a class of stochastic learning algorithms. The method applies whenever the algorithm generates an absolutely continuous distribution relative to some a-priori measure and the Radon Nikodym derivative has subgaussian concentration. Applications are bounds f...

Full description

Saved in:

Bibliographic Details
Main Author	Maurer, Andreas
Format	Journal Article
Language	English
Published	23.05.2024
Subjects	Computer Science - Learning
Online Access	Get full text
DOI	10.48550/arxiv.2405.14469

Cover

More Information
Summary:	The paper proves generalization results for a class of stochastic learning algorithms. The method applies whenever the algorithm generates an absolutely continuous distribution relative to some a-priori measure and the Radon Nikodym derivative has subgaussian concentration. Applications are bounds for the Gibbs algorithm and randomizations of stable deterministic algorithms as well as PAC-Bayesian bounds with data-dependent priors.
DOI:	10.48550/arxiv.2405.14469