Loss landscape Characterization of Neural Networks without Over-Parametrization
–Neural Information Processing Systems
In order to address this limitation, we propose a novel class of functions that can characterize the loss landscape of modern deep models without requiring extensive over-parametrization and can also include saddle points.
Neural Information Processing Systems
Oct-10-2025, 02:37:58 GMT
- Country:
- Africa > Senegal
- Kolda Region > Kolda (0.04)
- Asia
- Middle East > Jordan (0.04)
- Russia (0.04)
- Europe
- Germany
- Baden-Württemberg > Tübingen Region
- Tübingen (0.04)
- Bavaria > Lower Franconia
- Würzburg (0.04)
- Baden-Württemberg > Tübingen Region
- Netherlands > South Holland
- Dordrecht (0.04)
- Russia (0.04)
- Switzerland > Basel-City
- Basel (0.04)
- Germany
- Africa > Senegal
- Genre:
- Research Report > Experimental Study (1.00)
- Industry:
- Education (0.46)
- Technology: