Academic Year 2023-2024



Alberto Policriti
Unit Credits
Teaching Period
First Period
Course Type
Prerequisites. Since this is a foundational course, no prerequisites are required other than the computer science and mathematics maturity expected for students of the master’s degree course in Computer Science
Teaching Methods. There will be lectures and contributions in additional seminar form.
Verification of Learning. Verification of learning will take place through an oral exam and an in-depth study planned and prepared in collaboration with the teacher
More Information. The course will be held online
The main objective is to introduce the basic concepts, results, terminology and, above all, the mathematical tools commonly used in the field of neural networks today. The field is rapidly expanding and requires a firm understanding of the fundamental aspects and computational limits of the discipline. In this sense, a secure basic knowledge of the fundamental aspects of the formal tool that neural networks represent, is essential.
The course will be an introduction to the Neural Networks, with a specific focus on their history, their successes / defeats and their current status from both a theoretical and application point of view. In the initial part, biological neural networks will be briefly observed; we will then go on to illustrate what are the basic components of a neural network and how they are integrated into the model. We will then move on to studying the learning process in its main variants (unsupervised / supervised, reinforced, etc.) and the most popular network strategies / topologies. Following we will see a gallery of network models (Jordan, Elman, Hopfield, Radial-Basis, etc.) and, if time allows, we will discuss current themes in NN design: transformers, logical neural networks, fast/slow thinking models.
A Brief Introduction to Neural Networks – D. Kriesel

Introduction to Deep Learning E. Charniak

Neural Networks and Learning Machines S. Haykin

Shalev-Shwartz, Ben-David Understanding Machine Learning

Foundations of Data Science Blum A., Hopcroft J. and Kannan R.