A PAC-Bayesian Generalization Bound for Equivariant Networks.
Arash BehboodiGabriele CesaTaco S. CohenPublished in: NeurIPS (2022)
Keyphrases
- pac bayesian
- generalization bounds
- distribution free
- data dependent
- learning theory
- rademacher complexity
- vc dimension
- generalization ability
- upper bound
- model selection
- error bounds
- linear classifiers
- empirical risk minimization
- learning algorithm
- statistical learning theory
- large deviations
- worst case
- lower bound
- inductive inference
- decision trees