A singular Riemannian geometry approach to Deep Neural Networks II. Reconstruction of 1-D equivalence classes.
Alessandro BenfenatiAlessio MartaPublished in: CoRR (2021)
Keyphrases
- equivalence classes
- neural network
- equivalence class
- three dimensional
- information geometry
- bayesian network structures
- shape reconstruction
- pattern recognition
- geodesic distance
- directed acyclic graph
- image reconstruction
- euclidean space
- fuzzy logic
- high resolution
- artificial neural networks
- euclidean metric
- manifold learning
- affine invariant
- reconstruction method
- fisher information
- fine scale
- back propagation
- geometric structure
- shortest path