Presentations of the advanced topics in optimization
- Introduction
- Gradient descent and beyond. Part 1
- Stochastic approximation and sample average approximation
- Proximal methods
- Mirror descent + comparison with projected subgradient method
- Stochastic methods to train deep neural networks
- Catalyst and acceleration techniques
- Universal gradient methods
- Riemanien optimization
- Submodular optimization
- Convex optimization cheatsheet, not perfect, but still