MACHINE LEARNING METHODS FOR PHYSICS (D)

Mathematical-Methods
2 YEAR (Block D)
2 semester 8 CFU
(from Physics LM-17 )
Prof. Michele BUZZICOTTI A.Y. 2025-26
Code: 8067607
SSD: FIS/01
https://www.master-mass.eu/

 

  • PREREQUISITES: Basic concepts of Linear Algebra, Mathematical Analysis and Python Programming
  • OBJECTIVE: The lectures are thought to give a solid knowledge of the theoretical Machine Learning (ML) background. A special focus is given to the ML application for data analysis of physical systems. The students will also learn how to implement a typical ML model using the standard libraries in a Python environment.

 

COMPUTER VISION (2024-25)

COMPUTER VISION (2024-25)
2 YEAR II semester  6 CFU
Arianna Mencattini A.Y. 2023-24 (ex MEASUREMENT SYSTEMS FOR MECHATRONICS)
A.Y. 24-25
Code: 8039787
SSD: ING/INF/07

LEARNING OUTCOMES:
Learning of basic concepts in digital image processing and analysis as a novel measurement system in biomedical fields. The main algorithms will be illustrated, particularly devoted to the mechatronics fields.

KNOWLEDGE AND UNDERSTANDING:
The student acquires the knowledge related to the possibility to use an image analysis platform to monitor the dynamics of a given phenomenon and to extract quantitative information from digital images such as object localization and tracking in digital videos.

APPLYING KNOWLEDGE AND UNDERSTANDING:
The student acquires the capability to implement the algorithms in Matlab through dedicated lessons during the course to the aim of being able to autonomously develop new codes for the solution of specific problems in different application fields.

MAKING JUDGEMENTS:
The student must be able to integrate the basic knowledge provided with those deriving from the other courses such as probability, signal theory, and pattern recognition. some fundamentals of measurement systems as well as of basic metrological definitions will be provided in support of background knwoledge.

COMMUNICATION SKILLS:
The student solves a written test and develops a project in Matlab that illustrates during the oral exam. The project can be done in group to demonstrate working group capabilities.

LEARNING SKILLS:
Students will be able to read and understand scientific papers and books in English also to deepen some topics. In some cases, students will develop also experimental tests with time-lapse microscopy acquisition in department laboratory.

SYLLABUS

Fundamentals of metrology. Basic definitions: resolution, accuracy, precision, reproducibility and their impact over an image based measurement system . Image processing introduction. Image representation. Spatial and pixel resolution. Image restoration. Deconvolution. Deblurring. Image quality assessment. Image enhancement. Image filtering for smoothing and sharpening. Image segmentation: pixel based (otsu method), edge based, region based (region growing), model based (active contour, Hough transform), semantic segmentation. Morphological operators. Object recognition and image classification. Case study: defects detection, object tracking in biology, computer assisted diagnosis, facial expression in human computer interface.
Matlab exercises.

TEXTS

Digital image processing, Gonzalez and Woods, Prentice Hall, New York, 2002.

BiPM, I. E. C., IFCC, I., IUPAC, I., & ISO, O. (2012). The international vocabulary of metrology— basic and general concepts and associated terms (VIM). JCGM, 200, 2012.

ATTENDANCE

Although attendance is optional, it is strongly recommended to follow the lessons. The professor recommends the students to subscribe the course on the Delphi website. The teams platform will be used as a consequence to communicate with the Professor, ask for doubts, and download the materials used for the lessons.

Deep Learnig and applications (block C1-opt)

Deep Learning
2 YEAR II semester  6 CFU

Eugenio Martinelli
A.Y. 2024-25 new
Didatticaweb
Code:
SSD: ING-INF/01

PREREQUISITES

Basic knowledge of probability theory, signal theory, and pattern recognition.

FORMATIVE OBJECTIVES

LEARNING OUTCOMES:
Learning the basic concepts of deep learning algorithms. The main Machine Learning algorithms will be covered, followed by a focus on those related to deep learning, with particular emphasis on their application in the field of mechatronics.

KNOWLEDGE AND UNDERSTANDING:
The student acquires knowledge related to the field of Machine Learning, with particular reference to the ability to extract quantitative and qualitative information from images and videos and multivariate data and their subsequent processing for regression and classification tasks.

APPLYING KNOWLEDGE AND UNDERSTANDING:
The student acquires the capability to implement the algorithms in Matlab through dedicated lessons during the course to the aim of being able to autonomously develop new codes for the solution of specific problems in different application fields.

MAKING JUDGEMENTS:
The student must be able to integrate the basic knowledge provided with those deriving from the other courses, such as probability, signal theory, and pattern recognition.

COMMUNICATION SKILLS:
The student develops a project in Matlab that illustrates during the oral exam. The project can be done in groups to demonstrate working group capabilities.

LEARNING SKILLS:
Students will need to be able to read and understand scientific texts and articles in English for in-depth exploration of the topics covered. They should also independently expand their knowledge of the subject to include topics not directly addressed in the course, particularly those connected with the rapid technological developments in the field of Deep Learning and, more generally, in machine learning.

SYLLABUS

Today, deep neural networks surpass traditional hand-crafted algorithms and match human performance in various complex tasks, including image recognition, natural language processing, and prediction models. This course offers a comprehensive introduction to neural networks (NNs), covering traditional feedforward (FFNN) and recurrent (RNN) neural networks, as well as the most advanced deep-learning models like convolutional neural networks (CNN), Variational Autoencoders, and Diffusion models.

The primary objective of the course is to equip students with the theoretical knowledge and practical skills needed to understand and utilize neural networks (NN), while also familiarizing them with deep learning techniques for solving complex engineering challenges.
This goal is pursued in the course by:
• Describing the most important algorithms for NN training (e.g., backpropagation, adaptive gradient algorithms, etc.)
• Illustrating the best practices for successful training and using these models (e.g., dropout, data augmentation, etc.) in a practical session using a phyton environment.
• Providing an overview of the most successful Deep Learning architectures (e.g., convolutional networks, autoencoder for embedding, diffusion models, etc.)
• Providing an overview of the most successful applications with particular emphasis on models for solving visual recognition tasks.

TEXTS

Pattern recognition and machine learning, Christopher Bishop.

Deep Learning, Ian Goodfellow et al.

– slides of the professor

 

 

Multimedia Processing and Communication (block C-opt)

Multimedia Processing and Communication (block C-opt)
2 YEAR I semester  6 CFU
Tommaso Rossi

Cesare Roseti

ICT and Internet Engineering
A.Y. 2023-24
Code:
SSD: ING-INF/03

FORMATIVE OBJECTIVES

The course module provides an overview of the technologies involved in the multimedia application evolution from analogue to digital, from linear television to video on demand. To this aim, the module addresses the main TV standards, the TCP/IP protocols involved in modern streaming services, the network architectures and the different service modes.

PREREQUISITES: A good background in TCP/IP protocols.

SYLLABUS:

PARTE I – Digital TV standards, MPEG-2  and  Transport Stream, IP encapsulation over  DVB.

PARTE II – IP multicast, IGMP, IP multicast routing

PARTE III –  Transport protocols for IP multimedia applications; Video streaming applications and CDN, the multimedia protocol stack, RTP and RTCP, multimedia signalling protocols: RTSP, SDP and SIP, Key Performance Indicators.

PARTE IV -Adaptive Streaming over HTTP, MPEG-DASH, Support to multimedia applications over 5G.

Adaptive Systems (block C-opt) –> Identification and Neural Networks (24-25)

Adaptive Systems (block C-opt) –> Identification and Neural Networks (24-25)
2 YEAR II semester  6 CFU
Patrizio Tomei (4cfu)
Eugenio Martinelli (2cfu)
A.Y. 2023-24
SANTOSUOSSO Giovanni Luca A.Y. 2024-25
A.Y. 2025-26
(new name “Identification and Neural Networks”
Didatticaweb
Code: 80300088
SSD: ING-INF/04

Pre-requirement: The basics of systems theory and control are required.

LEARNING OUTCOMES: The course aims to provide the basic techniques for the design of predictors, filters, and adaptive controllers.

KNOWLEDGE AND UNDERSTANDING: Students must obtain a detailed understanding of design techniques with the help of MATLAB-SIMULINK to solve industrial problems of adaptive filtering, adaptive prediction, and adaptive control.

APPLYING KNOWLEDGE AND UNDERSTANDING: Students must be able to apply the project techniques learned in the course even in different industrial situations than those examined in the various phases of the course.

MAKING JUDGEMENTS: Students must be able to apply the appropriate design technique to the specific cases examined, choosing the most effective algorithms.

COMMUNICATION SKILLS: Students must be able to communicate using the terminology used for filtering, prediction, and adaptive control. They must also be able to provide logical and progressive exposures starting from the basics, from structural properties, from modeling to the design of algorithms, without requiring particular prerequisites. Students are believed to be able to understand the main results of a technical publication on the course topics. Guided individual projects (which include the use of Matlab-Simulink) require assiduous participation and exchange of ideas.

LEARNING SKILLS: Students must be able to identify the appropriate techniques and algorithms in real cases that arise in industrial applications. Furthermore, it is believed that students have the ability to modify the algorithms learned during the course in order to adapt them to particular situations under consideration.

Texts

Adaptive Filtering Prediction and Control, Graham C. Goodwin, Kwai Sang Sin, Dover Publications, 2009.