CMU Fall 2018 Convex Optimization lecture notes.
Notes
| # | Note | Status |
|---|---|---|
| 2 | Convexity I | draft |
| 3 | Convexity II | draft |
| 4 | Canonical Problem Forms | draft |
| 5 | Gradient Descent | draft |
| 6 | Subgradient | draft |
| 7 | Subgradient Methods | draft |
| 8 | Proximal Gradient Descent | draft |
| 9 | Stochastic Gradient Descent | draft |
| 10-11 | Duality | polished |
| 12 | Optimality Conditions for Constrained Optimization | draft |
| 13 | Duality Uses and Correspondents | polished |
| 14 | Newton Method | draft |
| 15 | Barrier Method | polished |
| 16 | Primal-Dual Interior-Point Methods | polished |
| 17 | Quasi-Newton Methods | polished |
| 19 | Numerical Linear Algebra | polished |
| 20 | Coordinate Descent | polished |
| 21-22 | Dual Decomposition and ADMM | polished |
| 23 | Frank-Wolfe Method | polished |
Concept Map
- Convexity and canonical problem forms
- First-order and proximal methods
- Duality, conjugates, and optimality conditions
- Newton methods, quasi-Newton methods, interior-point methods, and numerical linear algebra
- Dual decomposition, augmented Lagrangian methods, and ADMM
- Projection-free methods: Frank-Wolfe (conditional gradient)
References
Planned Notes
- Lecture 01 Introduction
- Lecture 18