The aims of the course concern theoretical, methodological, and practical issues related to the area of Soft Computing. In particular, the course: - is aimed at supplying basic knowledge necessary to analyse and evaluate the applicability of existing fuzzy and evolutionary solutions to specific problems; - is aimed at discussing methodological issues related to the application of fuzzy and evolutionary techniques to specific domains and contexts of application; - is aimed at presenting some specific technical and technological solutions for experimentation by the students.
Course Prerequisites
Mathematical-logical skills and Python basics acquired during the first year.
Teaching Methods
The course comprises usual lectures presented along with practical examples and case studies. Specific tools for the realization of the presented models and approaches will also be introduced in the classroom.
Slides and exercises will be made available online.
Assessment Methods
Learning assessment includes a written exam and an optional (according to teacher decision or student request) oral exam.
The written exam consists of open questions and exercises, equally distributed between the two parts of the course.
During the oral exam, besides discussing the contents of the written exam, some questions may be posed on all the subjects of the course.
An optional single/group project (1-3 members) will be proposed; a single possibility to carry out the project will be defined, plausibly close to the end of the course, with assignment due in the months of June/July. It could lead to extra points for the final evaluation, after a discussion of the project.
Texts
Kruse, R., Mostaghim, S., Borgelt, C., Braune, C., Steinbrecher, M. (2022). Introduction to Fuzzy Sets and Fuzzy Logics. In: Computational Intelligence. Texts in Computer Science. Springer, Cham. https://doi.org/10.1007/978-3-030-42227-1_15
Lecture notes and further reading provided during the semester.
Contents
FUZZY SYSTEM
- Basic notions: membership functions, set operations, fuzzy sets in uncertainty representation - How to define membership functions - Measures of Fuzziness - Hints to possibility theory - Hints to fuzzy logics - Fuzzy Control - Fuzzy Clustering (and its evaluation)
EVOLUTIONARY COMPUTING
- Minimal background on optimization, Biological inspiration - Components of Evolutionary Algorithms - Variation components: Representation, Mutation, and Recombination - Fitness, Selection, and Population Management - Variants: genetic algorithms, evolutionary programming, particle swarm optimization - Hints to parameters tuning and control - Hints to theoretical analysis
- Fuzzy Systems and Evolutionary computing hybridization
