Pattern Recognition and Machine Learning

Relations:

During a review of study plans, the course B4B33RPZ can be substituted for the course BE5B33RPZ.

It is not possible to register for the course BE5B33RPZ if the student is concurrently registered for or has already completed the course B4B33RPZ (mutually exclusive courses).

It is not possible to register for the course BE5B33RPZ if the student is concurrently registered for or has already completed the course A4B33RPZ (mutually exclusive courses).

In order to register for the course BE5B33RPZ, the student must have registered for the required number of courses in the group BEZBM no later than in the same semester.

It is not possible to register for the course BE5B33RPZ if the student is concurrently registered for or has previously completed the course B4B33RPZ (mutually exclusive courses).

Course guarantor:

Jiří Matas

Lecturer:

Ondřej Drbohlav, Jiří Matas

Tutor:

Ondřej Drbohlav, Jiří Matas, Lukáš Neumann, Jan Šochman

Supervisor:

Department of Cybernetics

Synopsis:

The basic formulations of the statistical decision problem are presented. The necessary knowledge about the (statistical) relationship between observations and classes of objects is acquired by learning on the raining set. The course covers both well-established and advanced classifier learning methods, as Perceptron, AdaBoost, Support Vector Machines, and Neural Nets.

This course is also part of the inter-university programme prg.ai Minor. It pools the best of AI education in Prague to provide students with a deeper and broader insight into the field of artificial intelligence. More information is available at https://prg.ai/minor.

Requirements:

Knowledge of linear algebra, mathematical analysis and

probability and statistics.

Syllabus of lectures:

1. Introduction. Basic notions. The Bayesian recognition problem

2. Non-Bayesian tasks

3. Parameter estimation of probabilistic models. Maximum likelihood method

4. Nearest neighbour method. Non-parametric density estimation.

5. Logistic regression

6. Classifier training. Linear classifier. Perceptron.

7. SVM classifier

8. Adaboost learning

9. Neural networks. Backpropagation

10. Cluster analysis, k-means method

11. EM (Expectation Maximization) algorithm.

12. Feature selection and extraction. PCA, LDA.

13. Decision trees.

Syllabus of tutorials:

You will implement a variety of learning and inference algorithms on simple pattern recognition tasks. Each week a new assignment is introduced at the beginning of the lab, and you are expected to complete the task during the submission period. The discussion at the beginning of the lab session will link the theory presented in the lectures to the practical task in the weekly assignments. The remaining time of the lab is devoted to individual interactions between students and teaching assistants.

1. Introduction, work with python, simple example

2. Bayesian decision task

3. Non-bayesian tasks - the minimax task

4. Non-parametrical estimates - parzen windows

5. MLE, MAP and Bayes parameter estimation

6. Logistic regression

7. Problem solving / exam questions

8. Linear classifier - perceptron

9. Support Vector Machine

10. AdaBoost

11. K-means clustering

12. Convolutional neural networks

13. Problem solving / exam questions

Study Objective:

To teach the student to formalize statistical decision making problems, to use machine learning techniques and to solve pattern recognition problems with the most popular classifiers (SVM, AdaBoost, neural net, nearest neighbour).

Study materials:

1.Duda, Hart, Stork: Pattern Classification, 2001.

2.Bishop: Pattern Recognition and Machine Learning, 2006.

3.Schlesinger, Hlavac: Ten Lectures on Statistical and Structural Pattern Recognition, 2002.

Note:

Further information:

https://cw.fel.cvut.cz/wiki/courses/BE5B33RPZ

Time-table for winter semester 2024/2025:

	06:00–08:0008:00–10:0010:00–12:0012:00–14:0014:00–16:0016:00–18:0018:00–20:0020:00–22:0022:00–24:00
Mon	roomKN:E-301 Matas J. Drbohlav O. 16:15–17:45 (lecture parallel1) Karlovo nám. room Drbohlav O. 18:00–19:30 (lecture parallel1 parallel nr.103)
Tue
Wed
Thu	roomKN:E-132 Neumann L. 12:45–14:15 (lecture parallel1 parallel nr.101) Karlovo nám. roomKN:E-126 Šochman J. 12:45–14:15 (lecture parallel1 parallel nr.102) Karlovo nám.
Fri

Time-table for summer semester 2024/2025:

Time-table is not available yet

The course is a part of the following study plans:

• Electrical Engineering and Computer Science (EECS) (compulsory elective course)
• Open Informatics - Computer Science 2016 (compulsory course in the program)
• Open Informatics - Internet of Things 2016 (compulsory course in the program)
• Open Informatics - Software 2016 (compulsory course in the program)
• Open Informatics - Computer Games and Graphics 2016 (compulsory course in the program)
• Open Informatics (compulsory course in the program)
• Open Informatics (compulsory elective course)
• Open Informatics - Artificial Intelligence and Computer Science 2018 (compulsory elective course)
• Open Informatics - Internet of Things 2018 (compulsory elective course)
• Open Informatics - Software 2018 (compulsory elective course)
• Open Informatics - Computer Games and Graphics 2018 (compulsory elective course)
• Electrical Engineering and Computer Science (EECS) (compulsory elective course)
• prg.ai/minor-tech (elective course)