Review for NeurIPS paper: Deep learning versus kernel learning: an empirical study of loss landscape geometry and the time evolution of the Neural Tangent Kernel

Open in new window

The document is on a web server that cannot be reached. Please try again later.