Talks (videos)
- M. Wyart, Elementary Excitations in Amorphous Materials (February 2017)
- M. Wyart, Theory for swap acceleration near the glass and jamming transition (September 2018)
- M. Wyart, Neural Tangent Kernel and over-parametrisation in deep learning (January 2019)
- S. Spigler, Loss Landscape and Performance in Deep Learning (January 2020)
- M. Wyart, Ecole de Physique des Houches, Loss Landscape and Performance in Deep Learning: varying the number of parameters and varying the number of training data (August 2020)
- M. Geiger, Euclidean equivariant neural networks (August 2020)
Slides
- M. Wyart, Scaling description of generalisation with number of parameters in deep learning (October 2019)
- S. Spigler, Jamming and deep learning (January 2020)
- S. Spigler, Learning curves of kernel methods. The curse of dimensionality? (March 2020)