PAC-Bayes-Chernoff bounds for unbounded losses
–Neural Information Processing Systems
In this direction, we provide a general bound under a new model-dependent assumption from which we obtain bounds based on parameter norms and log-Sobolev inequalities.
Neural Information Processing Systems
Oct-9-2025, 21:49:53 GMT
- Country:
- Asia > Japan
- Honshū > Kansai > Kyoto Prefecture > Kyoto (0.04)
- Europe
- Denmark > North Jutland
- Aalborg (0.04)
- Spain > Galicia
- Madrid (0.04)
- United Kingdom > England
- Oxfordshire > Oxford (0.04)
- Denmark > North Jutland
- North America > Canada
- Asia > Japan
- Genre:
- Research Report > Experimental Study (0.93)