Intelligenza artificiale LM 2010/2011

main

lab

pages

main • lab

Course materials

Course outline.
Lecture slides can be downloaded from AMS Campus.
Lab page

Additional teaching materials

The following papers are available only for teaching purposes.

AI history

Dartmouth summer research project proposal. The proposal for a project that has been the origin of AI.

Search

Search strategies exercises (in Italian). Credits Prof. Paola Mello.

CSP

A brief introduction to mainstream techniques of constraint satisfaction. R.Barták. Constraint propagation and backtracking-based search. CP Summer school 2005.
Incomplete depth-first search techniques: a short survey. R.Barták, Proceedings of CPDC 2004.
Where the Really Hard Problems Are. P.Cheesman et al., Proc. 12th IJCAI, 1991.
Phase transitions and the search problem. T.Hogg et al., Artificial Intelligence, n.81, 1996.

Metaheuristics

Introduction to metaheuristics. C.Blum, A.Roli. Metaheuristics in Combinatorial Optimization: Overview and Conceptual Comparison. ACM Computing Surveys, Vol.35, N.3, 2003. (preprint available)
A multi-agent architecture for metaheuristics. M.Milano, A.Roli. MAGMA: A Multiagent Architecture for Metaheuristics. IEEE Trans. on Systems, Man and Cybernetics - Part B, Vol.34, Issue 2, April 2004. (preprint available)
A formal model of local search. A.Roli. A note on a model of local search. Technical report TR/IRIDIA/2004/23.01, IRIDIA, Université Libre de Bruxelles, Belgium.
On the relation between complete and incomplete search. M.Milano, A.Roli. Proceedings of CPAIOR 2002.
An introduction to Hybrid metaheuristics (slides).
Large Neighbourhood Search Algorithms for the Founder Sequence Reconstruction Problem. A. Roli, S. Benedettini, T. Stuetzle and C. Blum. TR/IRIDIA/2010-012.
Comet source code of examples of stochastic local search algorithms for the N-queens problem.

Evolutionary computation

Book (free download, Creative Commons) on genetic programming. R. Poli, W.B. Langdon, N.F. McPhee, J. Koza. A Field Guide to Genetic Programming.

Swarm intelligence

M. Dorigo, E. Bonabeau, G. Theraulaz. Ant algorithms and stigmergy. Future Generation Computer Systems, Vol.16, n.9, 2000.
C. Blum. Ant colony optimization: Introduction and recent trends. Physics of Life Reviews, 2(4):353-373, 2005.
H. Labella, M. Dorigo, J.-L. Deneubourg. Division of Labour in a Group of Robots Inspired by Ants' Foraging Behaviour. Technical Report IRIDIA-TR-2004-13, IRIDIA, Université Libre de Bruxelles, Brussels, Belgium, 2005.

Logics

Alcune considerazioni sulla logica proposizionale. D. Palladino, 2003.
Introduction to SAT solving algorithms. C.P. Gomes, H. Kautz, A. Sabharwal and B. Selman. Satisfiability solvers. Book chapter draft, 2007.
Introduction to fuzzy system. B. Kosko and S. Isaka, Fuzzy Logic. Scientific American, July 1993.1.1.1 Software

Acknowledgements

I thank Prof. Paola Mello, Prof. Michela Milano and Prof. Giorgio Buttazzo for giving me the permission of using part of their teaching material, which was revised and emended for this course.

AI*IA - Italian association for artificial intelligence

The Italian association for artificial intelligence is a non profit association promoting AI research. Website: www.aixia.it L'Associazione Italiana per l'Intelligenza Artificiale è un'associazione non profit per la promozione dello studio e della ricerca nel campo dell'intelligenza artificiale. Offre ogni anno premi per tesi sull'Intelligenza Artificiale e borse di studio per la partecipazione ad eventi ai suoi iscritti. Per informazioni, gli studenti possono rivolgersi al docente di questo corso o ai referenti indicati sul sito dell'associazione (www.aixia.it).

learning outcomes

The course aims at providing students the fundamentals of Artificial Intelligence (AI). Students will be able to tackle search problems using classical AI algorithms, such as uninformed tree-search algorithms, Best First Greedy and A*, as well as constraint satisfaction and constrained optimization problems by means of backtracking search, consistency techniques, constraint programming and metaheuristics. Moreover, students will know the basics of evolutionary computation, swarm intelligence techniques and neural networks. As far as knowledge representation and automatic reasoning is concerned, students will learn how propositional and first order logics are used in expert systems and classical planning techniques.

course contents

Introduction
- Foundations of AI
- History of AI
- Application domains of AI
Problem solving
- Solving Problems by Searching
- Non-informed and informed Search and Exploration
- Constraint Satisfaction Problems (complete and incomplete techniques: standard backtracking, constraint propagation techniques, local search)
- Adversarial Search: two players games, games with uncertainty
Decision support systems and technologies
- Knowledge representation
- Inference in propositional and first order logic
- Reasoning
Planning
- Introduction to planning and scheduling problems
- Planning and scheduling solving techniques (basics)
Machine learning
- Reinforcement learning
- Evolutionary computation
- Neural networks

readings / bibliography

Russell, Norvig, "Artificial intelligence: A modern approach", Vol.1 and 2 (partially), second edition, Pearson/Prentice Hall

teaching methods

Lectures in classroom
Lectures in lab (AI software tools)

assessment methods

Written and oral examination.

teaching tools

Lecture slides (in English) and web resources
AI software tools