Table of contents
5 chapters · 11 sections
Optimisation Theory
- 1.1Convex and non-convex optimisation
- 1.2Stochastic gradient methods
- 1.3Implicit regularisation
Approximation Theory
- 2.1Universal approximation
- 2.2Depth vs width
Neural ODEs & Flows
- 3.1Continuous-depth models
- 3.2Normalising flows
Diffusion Models
- 4.1Score-based generative modelling
- 4.2SDE formulation
Symmetry & Geometry in Learning
- 5.1Equivariant networks
- 5.2Symmetry-driven machine learning