ID:
SCV0631
Durata (ore):
72
CFU:
9
SSD:
SISTEMI DI ELABORAZIONE DELLE INFORMAZIONI
Anno:
2024
Dati Generali
Periodo di attività
Primo Semestre (23/09/2024 - 20/12/2024)
Syllabus
Obiettivi Formativi
The course provides broad coverage of intelligent systems solving pattern recognition problems. Theoretical concepts in intelligent systems and techniques relevant to real-life applications will be illustrated.
The student will be able to:
1. Know the main objectives and areas of Artificial Intelligence, Machine Learning, and Pattern Recognition, with the ability to identify the potentialities of intelligent techniques and the relationships with other disciplines
2. Know the basic concepts of automated learning based on machine learning approaches and the conditions for their applicability
3. Know the most relevant feature extraction and selection techniques
4. Know statistical techniques and their limitations and strengths, with the ability to appropriately select the proper technique in specific contexts
5. Know basic principles of neural computing and their characteristics
6. Know Flat and Hierarchical Clustering with the ability to configure and apply these methods in specific contexts
7. Know performance metrics for learners
8. Know basic concepts of the following application domains: Image Classification, Text Categorization, Biomedical Data Analysis
9. Know how to program in a language for statistical computing and machine learning applications like R
It is also expected that students develop communicative skills through open discussion and autonomous assessment in the choice of the proper technique to solve problems of recognition and /or automatic classification of multidimensional data in several domains.
Students will acquire also knowledge of the relevant Machine learning and Pattern Recognition terminology.
The student will be able to:
1. Know the main objectives and areas of Artificial Intelligence, Machine Learning, and Pattern Recognition, with the ability to identify the potentialities of intelligent techniques and the relationships with other disciplines
2. Know the basic concepts of automated learning based on machine learning approaches and the conditions for their applicability
3. Know the most relevant feature extraction and selection techniques
4. Know statistical techniques and their limitations and strengths, with the ability to appropriately select the proper technique in specific contexts
5. Know basic principles of neural computing and their characteristics
6. Know Flat and Hierarchical Clustering with the ability to configure and apply these methods in specific contexts
7. Know performance metrics for learners
8. Know basic concepts of the following application domains: Image Classification, Text Categorization, Biomedical Data Analysis
9. Know how to program in a language for statistical computing and machine learning applications like R
It is also expected that students develop communicative skills through open discussion and autonomous assessment in the choice of the proper technique to solve problems of recognition and /or automatic classification of multidimensional data in several domains.
Students will acquire also knowledge of the relevant Machine learning and Pattern Recognition terminology.
Prerequisiti
The course assumes that students have a background acquired in a Bachelor's Degree in STEM disciplines. Students are expected to be familiar with basic Mathematics, Probability, and Statistics.
Metodi didattici
Lectures (72 hours)
The topics of the course are illustrated by means of (1) conceptual, formal descriptions, (2) their implementation via R code, and (3) the use of demos and online resources.
Constant interaction with the students and their involvement in open discussions are highly encouraged.
The topics of the course are illustrated by means of (1) conceptual, formal descriptions, (2) their implementation via R code, and (3) the use of demos and online resources.
Constant interaction with the students and their involvement in open discussions are highly encouraged.
Verifica Apprendimento
The students’ learning extent is assessed via a written test (duration: 2 hours) and an individual assignment, autonomously developed by each student individually.
The goal of the written test is to assess the learning degree and the understanding of the elements related to intelligent systems from both theoretical and application (on problems of limited complexity) points of view. Written tests normally consist of
- two exercises for the assessment of the student's understanding and knowledge of machine learning techniques: each exercise weighs about one-quarter of the grade of the written exam;
- four questions on the conceptual aspects: each exercise weighs about one-eighth of the grade of the written exam.
The assignment allows the students to use their skills and knowledge for the building and evaluation of machine learners by using the R language. The project presentation has the goal of assessing the students’ communication skills in two areas: 1) the students’ technical competencies and use of the correct terminology and 2) the students’ skills for communicating a complete and organized view of the work they carried out.
Individual judgment skills are evaluated based on the decisions made during the written exam and the assignment.
The grade of the written test is on a 0 to 30 scale. The written exam contributes 70% of the final mark, while the assignment accounts for the remaining 30%.
The goal of the written test is to assess the learning degree and the understanding of the elements related to intelligent systems from both theoretical and application (on problems of limited complexity) points of view. Written tests normally consist of
- two exercises for the assessment of the student's understanding and knowledge of machine learning techniques: each exercise weighs about one-quarter of the grade of the written exam;
- four questions on the conceptual aspects: each exercise weighs about one-eighth of the grade of the written exam.
The assignment allows the students to use their skills and knowledge for the building and evaluation of machine learners by using the R language. The project presentation has the goal of assessing the students’ communication skills in two areas: 1) the students’ technical competencies and use of the correct terminology and 2) the students’ skills for communicating a complete and organized view of the work they carried out.
Individual judgment skills are evaluated based on the decisions made during the written exam and the assignment.
The grade of the written test is on a 0 to 30 scale. The written exam contributes 70% of the final mark, while the assignment accounts for the remaining 30%.
Contenuti
The acquisition of knowledge and expected skills is developed throughout the entire course, which includes the topics listed below.
1) Introduction to Artificial Intelligence and Pattern Recognition: Historical Perspective, State of the Art of Methods and Applications (3 h - Course Objective 1)
2) Basic Mathematical and Statistical Concepts:
• Measurement Theory
• Matrix algebra
• Multivariable function analysis calculus: partial derivatives and gradient
• Relevant concepts of Probability and Statistics
• Relevant concepts of Information Theory
(6 h - Course Objective 2)
3) Design of a Supervised Classifier; Basic principles of learning by example; basic concepts of multidimensional pattern analysis
(5 h - Course Objective 2)
4) Feature Extraction and Selection:
• Principal Component Analysis
• Information Gain
• Statistical evaluation of features
• Selection Strategies
(6 h - Course Objective 3)
5) Fundamental Elements of Programming with R
(8 h - Course Objective 9)
6) Machine Learning Algorithms
• Ordinary Least Squares Regression
• Outliers and Robust Regression
• Regularization and Shrinkage in Regression: Ridge Regression, LASSO Regression, Elastic Net
• Minimum distance classifier
• Bayesian classification
• Maximum likelihood classifier
• K-Nearest Neighborhood classifier
• Parallelepiped Method
• Decision trees
• Ensemble models: boosting, bagging, stacking
• Support Vector Machine
• Imbalance, Hyperparameter tuning
• Performance metrics
• (32 h - Course Objectives 2, 3, 4, 7)
7) Neural Networks
• Introduction, taxonomy
• Basic principle of neural computing
• Feedforward Neural Models
• Application Examples
• Introduction to Deep Learning
(6 h - Course Objective 5)
8) Clustering
• Introduction to Clustering
• K-means Clustering algorithm
• Agglomerative Hierarchical Clustering: Single linkage, Complete linkage
(3 h - Course Objective 6)
9) Design of Intelligent Systems, Examples in Application Domains
(3 h - Course Objectives 1, 8)
1) Introduction to Artificial Intelligence and Pattern Recognition: Historical Perspective, State of the Art of Methods and Applications (3 h - Course Objective 1)
2) Basic Mathematical and Statistical Concepts:
• Measurement Theory
• Matrix algebra
• Multivariable function analysis calculus: partial derivatives and gradient
• Relevant concepts of Probability and Statistics
• Relevant concepts of Information Theory
(6 h - Course Objective 2)
3) Design of a Supervised Classifier; Basic principles of learning by example; basic concepts of multidimensional pattern analysis
(5 h - Course Objective 2)
4) Feature Extraction and Selection:
• Principal Component Analysis
• Information Gain
• Statistical evaluation of features
• Selection Strategies
(6 h - Course Objective 3)
5) Fundamental Elements of Programming with R
(8 h - Course Objective 9)
6) Machine Learning Algorithms
• Ordinary Least Squares Regression
• Outliers and Robust Regression
• Regularization and Shrinkage in Regression: Ridge Regression, LASSO Regression, Elastic Net
• Minimum distance classifier
• Bayesian classification
• Maximum likelihood classifier
• K-Nearest Neighborhood classifier
• Parallelepiped Method
• Decision trees
• Ensemble models: boosting, bagging, stacking
• Support Vector Machine
• Imbalance, Hyperparameter tuning
• Performance metrics
• (32 h - Course Objectives 2, 3, 4, 7)
7) Neural Networks
• Introduction, taxonomy
• Basic principle of neural computing
• Feedforward Neural Models
• Application Examples
• Introduction to Deep Learning
(6 h - Course Objective 5)
8) Clustering
• Introduction to Clustering
• K-means Clustering algorithm
• Agglomerative Hierarchical Clustering: Single linkage, Complete linkage
(3 h - Course Objective 6)
9) Design of Intelligent Systems, Examples in Application Domains
(3 h - Course Objectives 1, 8)
Lingua Insegnamento
INGLESE
Altre informazioni
During the period in which the course is held, the students can meet with the instructor on class days. During the remainder of the year, the students need to contact the instructor to set up an appointment by e-mail at sandro.morasca@uninsubria.it. The instructor responds only to e-mail messages sent from the official student.uninsubria.it e-mail accounts.
Corsi
Corsi
INFORMATICA
Laurea Magistrale
2 anni
No Results Found