A singular Riemannian geometry approach to Deep Neural Networks I. Theoretical foundations.
Alessandro BenfenatiAlessio MartaPublished in: CoRR (2022)
Keyphrases
- theoretical foundation
- neural network
- information geometry
- theoretical framework
- geodesic distance
- artificial neural networks
- pattern recognition
- three dimensional
- genetic algorithm
- recurrent neural networks
- information theory
- fuzzy logic
- neural network model
- back propagation
- multilayer perceptron
- manifold learning
- fault diagnosis
- self organizing maps
- shape analysis
- shape space
- tangent space
- multi layer
- riemannian manifolds
- deep learning
- high dimensional