Syllabus – Introduction to Optimisation
Prerequisites: basic linear algebra, multi-variate calculus, probability
Introduction to Optimisation
Motivating Examples (Network Applications)
Dynamic Optimization (Dynamic Programming)
Applications to optimal routing
- Fundamental Concepts in Convex Optimization
- Convex Sets and Functions
- Convex Optimization Problems:
- Existence of Solutions
- Optimality Conditions
- Projection Theorem and Separation Theorems
- Lagrangian Duality Theory
- Linear Programming Duality
- Slater Condition
- Karush-Khun-Tucker System
- Static Optimization (vector-space methods)
- Primal Algorithms
- Simplex Algorithm
- Gradient Projection Algorithm
- Dual Algorithms (differentiable case only)
- Primal Algorithms
- Classical Network Flow Problems
- Shortest Path
- Max-Flow Min-Cut
- Distributed Network Optimization
- Kelly’s Utility Maximization Framework
- Examples of routing, congestion control, rate allocation
- Fundamental Concepts and Bellman’s Equation
- Stochastic Shortest Path
- Infinite Horizon Problems
- Discounted Cost
- Average Cost
- Dijkstra’s Shortest Path Algorithm
- Max-Throughput (Tassiulas, Stolyar, and Neely)
- Min-Delay Model (Gallagher’s data routing)