Artificial Intelligence
People
Course director
Mele U. J.
Assistant
Description
COURSE OBJECTIVES
Meta-Heuristics algorithms learning and testing
COURSE DESCRIPTION
The aim of the course is to present the most modern techniques for solving complex problems. We focus on state of the art meta-heuristic for continuous function and combinatorial optimization: among the methods we deepen simulated annealing, genetic algorithms, variable neighborhood search and ant colony optimization. Gaming with two players will be also discussed as the integration between meta-heuristics and machine learning. Students are asked to implement and test some of these techniques.
LEARNING METHODS
- Front lecture, text reading, code development
- For each topic you receive a paper to read and to support to your studies
Simulated Annealing
SA extension ILS: iterated local search
Genetic Algorithms
GA variant PBIL population-based incremental learning
Ant Colony Optimization
ACS for vehicle routing problems
Tabu search
EXAMINATION INFORMATION
Writen test and code evaluation
REFERENCES
- Artificial Intelligence: A Modern Approach (3rd Edition) by Stuart Russell, Peter Norvig
- Artificial Intelligence, third edition, P.H. Winston, Addison-Wesley
- Genetic Algorithms in Search, Optimisation, and Machine Learning, Goldberg, Addison-Wesley, MA
- Ant Colony Optimization By Marco Dorigo and Thomas Stützle A Bradford Book
in addition the seven state of the art papers
Course Material in English will be provided to the students
Education
- Master of Science in Artificial Intelligence, Foundation course, 2nd year
- Master of Science in Computational Science, Elective course, Lecture, 2nd year
- Master of Science in Computational Science, Elective course, 1st year
- Master of Science in Informatics, Elective course, 1st year
- Master of Science in Informatics, Elective course, 2nd year