What is it about?

This work is about a theoretical understanding of how the output of a neural network changes as one increases the depth of the network. In particular, such progression is interpreted as a time evolution and we derive exact dynamics for such quantities.

Featured Image

Read the Original

This page is a summary of: The Mori–Zwanzig formulation of deep learning, Research in the Mathematical Sciences, May 2023, Springer Science + Business Media,
DOI: 10.1007/s40687-023-00390-2.
You can read the full text:

Read

Contributors

The following have contributed to this page