Academic Year 2023-2024

ARTIFICIAL INTELLIGENCE FOR MULTIMEDIA

Teachers

Lauro Snidaro
Unit Credits
9
Teaching Period
First Period
Course Type
Characterizing
Prerequisites. Requirements: Basic programming skills required. Knowledge of an object-oriented programming language is a plus.
Teaching Methods. The entire course will be held in the laboratory, thus allowing immediate hands-on practice on the concepts acquired.
Verification of Learning. The exam will be based on a project and oral discussion.

The objective of the evaluation will be to assess:

– the knowledge and understanding acquired through questions on the fundamental concepts of Deep Learning and Generative Learning.

– skills in applying acquired knowledge and use of correct terminology through a discussion on the presented project

EVALUATION OF LEARNING OUTCOMES

30, 30 cum laude

EXCELLENT

The student:

demonstrates a comprehensive and detailed knowledge of the subject: knows the relevant content, uses correct terminology, identifies and explains the main concepts, integrates personal insights into their synthesis;

28-29

VERY GOOD

The student:

demonstrates a deep knowledge of the subject: knows the relevant content, uses correct terminology, identifies and explains most of the main concepts;

25-27

GOOD

The student:

demonstrates a broad knowledge of the subject: knows, although not fully explained, the relevant content; uses terminology, but not always precisely; identifies key concepts but fails to explain them fully or accurately.

22-24

FAIR

The student:

demonstrates an acceptable knowledge of the subject: knows the majority of the content but shows gaps, exhibits some confusion in certain important but non-essential concepts;

18-21

SUFFICIENT

The student:

demonstrates a limited knowledge of the subject: knows the most relevant content but exhibits numerous gaps, identifies a good portion of the key concepts but cannot explain them fully and accurately;

INSUFFICIENT

The student:

demonstrates poor and fragmented knowledge of the subject: does not know the essential content, showing extensive gaps, and does not identify the key concepts.

More Information. ADDITIONAL TEACHING MATERIAL:

Slides, video lessons. All material is available through elearning.uniud.it

THESES:

Theses are available. Contact the instructor.

Objectives
https://www.uniud.it/it/didattica/info-didattiche/regolamento-didattico-del-corso/LM-comunicazione-multimediale-tecnologie-informazione/all-B2
Contents
Part I: Fundamentals

Neural networks

Introduction to Deep Learning

Convolutional Networks

Introduction to Generative Models

Variational Autoencoders

Generative Adversarial Networks

Parte II: Applications

Generation of artificial images

Neural Style Transfer

Text generation

Composing music

Texts
Generative Deep Learning: Teaching Machines to Paint, Write, Compose, and Play, O’Reilly, 2019.

Rashid, Tariq. Make your own neural network: a gentle journey through the mathematics of neural networks, and making your own using the Python computer language. CreateSpace Independent Publ., 2016.

Goodfellow, Ian, et al. Deep learning. Vol. 1. Cambridge: MIT press, 2016.

Deep Learning with Keras: Implementing deep learning models and neural networks with the power of Python, Packt, 2017.