1. komponenta

Lecture type	Total
Lectures	45

* Load is given in academic hour (1 academic hour = 45 minutes)

COURSE AIMS AND OBJECTIVES: Introduction to the classical meta-heuristic methods and their usage for solving hard combinatorial problems.

COURSE DESCRIPTION AND SYLLABUS:
1. Introduction. Why are some problems difficult to solve. The size of the search space. Modeling the problem. Constraints.
2. Basic concepts. The objective. The evaluation function. Defining a search problem. Neighborhoods and local optima.
3. Traditional methods. Local and global search for SAT problem, the traveling salesman problem (TSP) and nonlinear programming problem (NLP). Linear programming. Greedy algorithms for SAT, TSP and NLP. Divide and Conquer. Dynamic programming.
4. Designing evolutionary algorithms. Representation. Evaluation function. Initialization. Selection. Mutation and crossover operators.
5. Escaping local optima. Simulated annealing. Tabu search.
6. Algorithm Analysis. Tuning the algorithm to the problem. Parameter control in evolutionary algorithms. Parent selection. Population. Genetic operators and their probabilities.

1. Z. Michalewicz, D. B. Fogel: How to Solve It: Modern Heuristic
2. J. Hromkovic: Algorithmics for Hard Computing Problems

Izborni predmet 1, 2 - Regular study - Computer Science and Mathematics

Izborni predmet 1, 2 - Regular study - Computer Science and Mathematics

Izborni predmet 3, 4, 5, 6 - Regular study - Computer Science and Mathematics

Izborni predmet 3, 4, 5, 6 - Regular study - Computer Science and Mathematics