Deep Learning

Module Core Models and Methods

Knowledge and Understanding: At the end of the course, students will have an in-depth understanding of basic Deep Learning models and methods, exploring techniques for tackling classical problems that underlie more advanced methodologies and applications. They will understand neural models, techniques for training neural models, convolutional neural networks, sequential learning models, and Bayesian learning models, developing a critical understanding of their theoretical and practical foundations.

Applying knowledge and understanding: Students will be able to apply basic deep learning models to classical problems (e.g., learning representations from labeled data, regression, classification, regularization, transfer learning). They will have the skills to implement and optimize models for heterogeneous data, work with neural architectures, and apply learning and optimization techniques such as gradient descent and back-propagation. They will also be able to understand and apply the principles of sequential learning and Bayesian inference.

Making judgments: Students will develop strong independent judgment in evaluating and selecting basic Deep Learning models and methods. They will be able to critically analyze the scientific literature, identifying the strengths and weaknesses of solutions and proposing innovative approaches. They will acquire the ability to make informed decisions about the applicability and effectiveness of the techniques presented in the course in real-world contexts, considering practical implications.

Communication skills: Students will have the expertise to discuss basic concepts related to Deep Learning models. They will be able to present projects and obtained results, using appropriate technical language. Communication skills will include the ability to critically discuss, collaborate in multidisciplinary groups, and draft high-quality technical documentation and scientific reports.

Learning skills: Students will develop a strong capacity for independent learning which is essential for staying at the forefront of the Deep Learning field. They will proactively identify and assimilate new theories, algorithms, and technologies, independently leveraging resources such as research articles, tutorials, and software libraries. This skill will prepare them to pursue more advanced paths, addressing future challenges in artificial intelligence.

The course takes place through lectures and laboratory activities, which will allow the practical application of theoretical concepts with exercises and case studies.

If the course is taught in mixed or distance mode, the necessary changes may be introduced with respect to what was previously stated, in order to comply with the program provided and reported in the syllabus.

Solid foundation of Machine Learning (linear/Logistic regression, Kernel Machines, Ensemble Methods) and familiarity with scientific programming in Python.

Attendance to the lessons is not mandatory but strongly recommended.

The course "Deep Learning: Core Models and Methods" explores the current deep learning technologies, providing the foundations. It begins with basic supervised learning techniques for solving classic regression and classification problems, such as Multilayer Perceptiron. It then addresses optimization and overfitting problems for deep learning models and convolutional neural architectures (CNNs). Transfer learning techniques will be presented, along with recurrent neural architectures useful for processing sequential data (e.g., series, videos), and an introduction to generative models. Modern architectures for solving state-of-the-art problems (e.g., Transformers) and Bayesian causal models will be discussed. Some lectures will be dedicated to the design of industrial applications.

The exam of the "Deep Learning" course consists of two individual exams, associated with the two modules. An individual grade is assigned to each exam. The grade of the overall exam is obtained as the average of the marks associated with the two modules "Core models and methods" and "Advanced models and methods". 

The exam of the "Core models and methods" module consists of a written test and an individual or group project, generally consisting of the implementation and analysis of an advanced Deep Learning model on a real dataset. The written test is aimed at verifying the theoretical knowledge of the topics of the course, as well as the expressive skills and language properties of the students. The written test generally consists of three open-ended questions. The project will assess the application of advanced knowledge, autonomy, quality of implementation and analysis. The project must be carried out after the written test. Each of the two tests is assigned a score out of thirty and the final grade assigned to the module is obtained by means of the arithmetic average of the two marks.

The exam of the module "Advanced models and methods" consists of a written test and an individual or group project, generally consisting of the implementation and analysis of an advanced Deep Learning model on a real dataset. The written test is aimed at verifying the theoretical knowledge of the topics of the course, as well as the expressive skills and language properties of the students. The written test generally consists of a set of multiple-choice and open-ended questions. The project will assess the application of advanced knowledge, autonomy, quality of implementation and analysis. The project must be carried out after the written test. Each of the two tests is assigned a score out of thirty and the final grade assigned to the module is obtained by means of the arithmetic average of the two marks.

The tests of the "Core models and methods" module must be taken before the tests of the "Advanced models and methods" module.

The following criteria will normally be followed for the attribution of the final grade:

·       Not approved: the student has not acquired the basic concepts and is not able to carry out the project.

·       18-23: the student demonstrates minimum mastery of advanced concepts; project with significant deficiencies.

·       24-27: the student demonstrates good mastery of advanced content; well-structured project and adequate analysis.

·       28-30 laude: the student has acquired all the advanced contents; Excellent and innovative design, complete, critical and error-free analysis, demonstrating full autonomy.

Students with disabilities and/or SLD must contact the teacher, the CInAP contact person of the DMI (Prof. Daniele) and the CInAP sufficiently in advance to take advantage of the appropriate compensatory measures. Learning assessments may also be conducted electronically, should circumstances require it.

-       Discuss the MLP architecture and possible loss functions for classification and regression problems.

-       Illustrate the backpropagation algorithm, including its pseudocode.

-       Formalize and discuss a Deep Sequential Learning architecture.

-       Discuss the principles of Bayesian Causal Learning algorithms.

-       Discuss convolutional architectures and transfer learning techniques.

It should be noted that these questions are purely indicative: the questions proposed for the examination may diverge, even significantly.