Objectives:
The course presents the theoretical foundations, the methodologies, and the technologies of artificial intelligence for information and knowledge processing. In particular, the course will present the main approaches: neural networks, fuzzy systems, and evolutionary computing.
Competences:
Theoretical foundation and practical use of the main computational intelligence approaches (neural networks, fuzzy systems, and evolutionary computing).
Program:
- Neural networks: Definitions. Neurons: structures, perceptrons, Multi-layered feed-forward networks. Self-organizing Maps. Hopfield networks. Radial Basis Functions networks, Support Vector Machines. Feed-back networks. Deep Learning networks. Learning: supervised, unsupervised. Performance. Optimization. Classification and clustering. Prediction. Function approximation.
- Fuzzy logic and systems: Fuzzy sets. Membership functions. Fuzzy rules. Defuzzification. Fuzzy reasoning. Fuzzy systems. Rough sets. Performance.
- Evolutionary computing: Genomic representation. Fitness functions. Selection. Genetic algorithms. Genetic programming. Evolutionary programming. Evolutionary strategies. Differential evolution. Swarm intelligence. Artificial immune systems.
Suggested background knowledge:
Fundamental concepts of computer science, computer programming, mathematics (discrete and continuous).
Textbook:
R. Kruse, C. Borgelt, C. Braune, S. Mostaghim, M.
Steinbrecher
Computational Intelligence: A Methodological Introduction
Springer, 2016
Exam:
Written test
- 3 essay questions on all topics covered by the course
- high-level questions
- sufficient grade requires sufficient answers to all questions
- maximum time: 2h
Contacts:
Prof. Angelo Genovese
email: angelo.genovese@unimi.it
tel: 02-503-16249
office: Dept. Computer Science, via Celoria 18, Milano, 6th floor, room 6002
meeting: by appointment taken by email
- Docente titolare: Angelo Genovese