Syllabus
- Local search, Hill climbing, Simulated annealing
- Population methods, e.g. Genetic algorithms
- Problem instance reduction, Large neighborhood search
- Hybrid methods: Lamarckian vs. Baldwinian learning, examples
- Surrogate models
- Applications, e.g. Minimum Common String Partition, Minimum Weight Dominating Set Problem, Arc Routing Problems, Public Transportation
The course is taught bi-yearly, alternating with the course Large-scale optimization: Exact methods (NOPT059).
Annotation
Lecture on heuristic optimization algorithms based on Convex
Optimization and Artificial Intelligence for solving real-life problems.