Pages that link to "Item:Q5854115"
From MaRDI portal
The following pages link to Comparing dynamics: deep neural networks versus glassy systems (Q5854115):
Displaying 17 items.
- An emergent autonomous flow for mean-field spin glasses (Q2032429) (← links)
- A spin glass model for the loss surfaces of generative adversarial networks (Q2076047) (← links)
- Landscape and training regimes in deep learning (Q2231925) (← links)
- The loss surfaces of neural networks with general activation functions (Q3382363) (← links)
- Maximal relevance and optimal learning machines (Q4992314) (← links)
- An analytic theory of shallow networks dynamics for hinge loss classification* (Q5020044) (← links)
- (Q5053228) (← links)
- Gradient descent dynamics and the jamming transition in infinite dimensions (Q5057865) (← links)
- Archetypal landscapes for deep neural networks (Q5073145) (← links)
- Ensemble Inference Methods for Models With Noisy and Expensive Likelihoods (Q5090110) (← links)
- Scaling description of generalization with number of parameters in deep learning (Q5856249) (← links)
- Mean-field dynamics of infinite-dimensional particle systems: global shear versus random local forcing (Q5857569) (← links)
- Universal characteristics of deep neural network loss surfaces from random matrix theory (Q5878969) (← links)
- Black holes and the loss landscape in machine learning (Q6061784) (← links)
- An introduction to machine learning: a perspective from statistical physics (Q6140146) (← links)
- On weak ergodicity breaking in mean-field spin glasses (Q6596499) (← links)
- Tradeoff of generalization error in unsupervised learning (Q6607283) (← links)
