OVERVIEW

Computational Intelligence constitutes a repertoire of Artificial Intelligence predictive methodologies build on data and on domain  knowledge, which are part of the background of the strategic engineer.

LEARNING OUTCOMES

Neural networks; fuzzy logic systems; evolutionary computing; swarm intelligence; neuro-fuzzy and fuzzy neural systems; hybrid intelligent systems, machine learning; classification, regression learning, clustering

AIMS AND LEARNING OUTCOMES

The course presents a systematic introduction to the foundations and the applications of Computational Intelligence models which are advanced data processing  methods of Artificial Intelligence inspired by natural systems and that encompass artificial neural networks, fuzzy logic systems, evolutionary calculus, swarm intelligence and machine learning. The most relevant topics, such as classification and regression, will be addressed both from a theoretical point of view and through practical programming exercises and homework using the Python language.

PREREQUISITES

The Course does not require specific prerequisites and includes all the necessary elements and references. The basic knowledge in mathematics, statistics acquired in previous studies, and programming skills in Python will be useful for improving the learning curve and student performance. An introduction to programming in Python is provided by the seminar W35: Programming  (Programming and Code Development Foundations)

Teaching methods

1 Lecture of 4 hours in a row per week for 10 weeks including frontal lectures, Class exercises and home-works.

SYLLABUS/CONTENT

Optimization; Machine Learning; Regression; Classification; Bayesian Decision Theory; Parametric Classification; Intro to clustering; Fuzzy Sets; Fuzzy Clustering;  Kernel Clustering; Spectral Clustering; Networks' Analysis; Neural Networks; Support Vector Machines;  Multi-Layer Perceptrons; Fuzzy Systems; Deep Learning;  Ensembles; Genetic Algorithms;  Evolution Strategies; Particle Swarm Optimization; Multi-Objective Genetic Algorithms; Multimodal Medical Volumes Segmentation; Seminars by companies operating in AI; Demos; Homeworks.

RECOMMENDED READING/BIBLIOGRAPHY

• Textbook: Andries P. Engelbrecht: Computational Intelligence - An introduction, Wiley, 2007.

• Selection of relevant journal papers

• Lecture notes / slides 

Exam Board

FRANCESCO MASULLI (President)

ALBERTO CABRI

AGOSTINO BRUZZONE

STEFANO ROVETTA (President Substitute)

Teaching methods

1 Lecture of 4 hours in a row per week for 10 weeks including frontal lectures, Class exercises and home-works.

LESSONS START

Spring Semester

Exam description

Homeworks and oral exam

Exam schedule