Geometric compression of invariant manifolds in neural nets

By J. Paccolat, Leonardo Petrini, Mario Geiger, Kevin Tyloo, Matthieu Wyart. 2020

14 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Yoshua Bengio
2. Mario Geiger
3. Stefano Spigler
4. Simon Lacoste-Julien
5. Sergey Levine
6. R. Devon Hjelm
7. Ruslan Salakhutdinov
8. Quoc V. Le

Citations in the corpus

• Disentangling feature and lazy training in deep neural networks (2019, methodology): “In a recent work (Paccolat et al 2020) we performed that comparison as p varies for a CNN trained on MNIST.”
• Implicit Regularization via Neural Feature Alignment (2020, result): “Independent concurrent works highlight alignment and compression phenomena similar to the one we study here (Kopitkov & Indelman, 2020; Paccolat et al., 2020).”