ARTIFICIAL INTELLIGENCE LABORATORY
cod. 1009073

Academic year 2022/23
2° year of course - First semester
Professor
Vincenzo BONNICI
Academic discipline
Informatica (INF/01)
Field
Discipline informatiche
Type of training activity
Characterising
48 hours
of face-to-face activities
6 credits
hub:
course unit
in ITALIAN

Learning objectives

The course provides the foundation for modeling and implementing algorithms and systems for solving artificial intelligence problems.
The course includes theoretical lessons that introduce the topics of machine learning, the search for solutions from an algorithmic point of view and the analysis of sequences, topics which are then treated from a practical and applicative point of view.

With reference to the Dublin indicators:

Knowledge and understanding
The course introduces the concepts related to the development, application and validation of artificial intelligence algorithms for machine learning and the search for solutions. In particular, supervised and unsupervised machine learning algorithms, search algorithms and sequence analysis algorithms are covered.

Ability to apply knowledge and understanding
The theoretical and practical knowledge presented is introduced in perspective both of the correct application of state-of-the-art algorithms and interpretation of the results obtained, and in understanding their motivations for the development of new computational solutions.

Judgment autonomy
A critical approach on the use and understanding of current tools for solving artificial intelligence problems is among the main objectives of the course. In particular, identify the correct methodology and evaluate its impact on the specific application case.

Communication skills
The discussions on the different methods to solve the proposed problems allow to improve communication skills through the development of a project on topics established by the teacher and which is tackled individually or in groups. The results of the project are then presented to the teacher.

Learning ability
The independent use of external resources and the consultation of existing scientific literature and practical tools allows for the development of independent learning skills. The student acquires the ability to adapt to the problem and to apply the most suitable models for the resolution.

Prerequisites

- - -

Course unit content

The course provides the foundation for modeling, implementing and applying algorithms for solving artificial intelligence problems.
The course includes theoretical lessons that introduce specific topics not covered in other courses e
related to machine learning, the search for algorithmic solutions and the design of systems that exploit artificial intelligence.
In particular, supervised and unsupervised machine learning algorithms, search algorithms and sequence analysis are covered.
The main notions on practical tools, in particular Python and related libraries, are provided during the course.

Full programme

Python for exploratory data analysis and machine learning.
Feature seleciton.
Dimensionality reduction.
Classification and clusterizzation.
Data set imbalance.
Introduction to regression methods.
Dynamic programming, Time warping, Sequence alignment.
Construction fo syntetic benchmarks.
Development of search algoritms.

Bibliography

Stuart, Norvig. Artificial intelligence: a modern approach. Pearson, 2016

Teaching methods

Classes with lessons and laboratory sessions. Group or single-student project.

Assessment methods and criteria

Oral exam with project.

Other information

- - -

2030 agenda goals for sustainable development

- - -