Semi-flat minima and saddle points by embedding neural networks to overparameterization
Kenji Fukumizu, Shoichiro Yamaguchi, Yoh-ichi Mototake, Mirai Tanaka
–Neural Information Processing Systems
We theoretically study the landscape of the training error for neural networks in overparameterized cases.
Neural Information Processing Systems
Aug-19-2025, 22:33:59 GMT
- Country:
- Asia > Japan
- Honshū > Kantō > Tokyo Metropolis Prefecture > Tokyo (0.04)
- North America
- Canada (0.04)
- United States > Florida
- Broward County > Fort Lauderdale (0.04)
- Asia > Japan
- Genre:
- Research Report > New Finding (0.46)
- Technology: