Pages that link to "Item:Q5857444"
From MaRDI portal
The following pages link to Disentangling feature and lazy training in deep neural networks (Q5857444):
Displaying 12 items.
- Linearized two-layers neural networks in high dimension (Q2039801) (← links)
- Learning what we don't care about: anti-training with sacrificial functions (Q2215057) (← links)
- Landscape and training regimes in deep learning (Q2231925) (← links)
- Geometric compression of invariant manifolds in neural networks (Q3382321) (← links)
- (Q4998974) (← links)
- An analytic theory of shallow networks dynamics for hinge loss classification* (Q5020044) (← links)
- Relative stability toward diffeomorphisms indicates performance in deep nets* (Q5055429) (← links)
- Harmonic analysis of network systems via kernels and their boundary realizations (Q6160662) (← links)
- Infinite-width limit of deep linear neural networks (Q6587580) (← links)
- Learning sparse features can lead to overfitting in neural networks (Q6611436) (← links)
- Self-consistent dynamical field theory of kernel evolution in wide neural networks (Q6611447) (← links)
- Redundant representations help generalization in wide neural networks (Q6611451) (← links)
