Insegnamento a.a. 2025-2026

The course is organized along the following themes:

• Recap of Machine Learning (ML) fundamentals.
• Introduction to Neural Networks and the connectionist paradigm: from the perceptron to Multi-Layer Perceptrons (MLPs), universality theorems, the backpropagation algorithm, and principles of Neural Network design.
• The rise of Deep Learning: Convolutional Neural Networks (CNNs), regularization techniques, and residual connections. Basics of Recurrent Neural Networks (RNNs), attention mechanisms, and Transformers.
• Introduction to Natural Language Processing (NLP): text preprocessing, static and contextual word embeddings, language modelling, and neural approaches to text processing—from neural machine translation to modern large language models (LLMs).

• Understand the motivations, historical origins, and evolution of Deep Learning.
• Recognize the strengths, limitations, and appropriate applications of neural networks.
• Grasp the key architectural and design principles underlying modern Deep Learning systems.
• Understand the fundamental concepts of Natural Language Processing (NLP) and the technical aspects of applying modern Deep Learning methods to NLP tasks.

• Design, implement, and evaluate Deep Learning algorithms for a range of tasks.
• Translate research concepts and ideas into practical implementations and real-world applications.
• Employ advanced Deep Learning frameworks and platforms effectively.

• Lectures
• Individual works / Assignments

• Core lectures: Face-to-face lectures introducing students to the core concepts of the course.
• Tutorial sessions: for deeper exploration of advanced topics.
• Practical exercise sessions: focused on coding, hands-on assignments, and in-class implementation.

Some individual assignments during the semester (30% of the final grade) + an end of semester reduced written exam (70% of the final grade).

A single full written exam (100% of the final grade).

• Slides and notes will be provided throughout the course. Within these resources, additional readings will be provided to students to expand on more advanced topics.
• Books:
  • Bishop, Chris, and Bishop, Hugh. Deep Learning: Foundations and Concepts, Springer, 2024
  • Prince, Simon J.D. "Understanding Deep Learning". MIT press (2023)
  • Goodfellow, Ian, et al. Deep learning. Vol. 1. No. 2. Cambridge: MIT press, 2016.