Connections between physics, mathematics, and deep learning
Starting from Fermat’s principle of least action, which governs classical and quantum mechanics and from the theory of exterior differential forms, which governs the geometry of curved manifolds, we show how to derive the equations governing neural networks in an intrinsic, coordinate-invariant way...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Andromeda Publishing and Academic Services
2019-09-01
|
| Series: | Letters in High Energy Physics |
| Online Access: | http://journals.andromedapublisher.com/index.php/LHEP/article/view/110 |