Uniform Generalization Bounds on Data-Dependent Hypothesis Sets via PAC-Bayesian Theory on Random Sets.
Benjamin DupuisPaul ViallardGeorge DeligiannidisUmut SimsekliPublished in: CoRR (2024)
Keyphrases
- data dependent
- generalization bounds
- pac bayesian
- large deviations
- rademacher complexity
- learning theory
- risk bounds
- generalization ability
- statistical learning theory
- empirical risk minimization
- distribution free
- learning machines
- linear classifiers
- model selection
- machine learning
- uniform convergence
- vc dimension