Simplifying Momentum-based Positive-definite Submanifold Optimization with Applications to Deep Learning.
Wu LinValentin DuruisseauxMelvin LeokFrank NielsenMohammad Emtiyaz KhanMark SchmidtPublished in: ICML (2023)
Keyphrases
- deep learning
- positive definite
- kernel machines
- unsupervised learning
- kernel function
- covariance matrix
- mental models
- geometric structure
- kernel matrix
- decision trees
- semi parametric
- weakly supervised
- sample size
- constrained optimization
- denoising
- diffusion tensor
- semi supervised
- pattern recognition
- support vector
- data mining