COURSE DESCRIPTION
An introductory course to the theory and practice of machine learning, recognition, and artificial intelligence
In the accreditation valid from 2021-22, this course replaced the course ROZ2. It is not an equivalent, but a modernized replacement, reflecting current developments in artificial intelligence and modified to be accessible to undergraduate students. The direct continuation is then the SU2 course.
COURSE SCHEDULE [2023/2024]
The lectures are held in the building of UTIA CAS, Pod vodárenskou věží 4, Prague 8. Transportation: metro C, stop Ládví. Meet students in the lobby 10 minutes before the start of the class.
The exercises take place online on Zoom.
Tentative lecture time: Wednesday, 10:00 - 12:00.
All students, who signed up for the course in KOS, will receive an email with time confirmation and other instructions.
COURSE OUTLINE
Introduction to artificial intelligence, basic concepts
Decision-making and recognition as an optimization problem
Theory of feature recognition, "handcrafted" and "learned" features
Intuitive classifiers with learning - NN classifier, k-NN classifier, linear classifier
Support Vector Machines (SVM) classifiers, simple neural networks
Bayesian approach to recognition
Classification without learning - cluster analysis in the posterior of features, iterative and hierarchical methods
Dimensionality reduction of the feature space, PCA, discriminant analysis
COURSE HANDOUTS
The files contain most of the slides used. This is not the complete lecture content; the knowledge recorded in the files is insufficient to pass the exam!
Always read the relevant part of [A] and, if possible, at least indicatively from [1] (Lectures 2,3,4) or [3] (Lecture 5).
COMPULSORY LITERATURE
RECOMMENDED LITERATURE
[1] Duda R.O. et al., Pattern Classification, (2nd ed.), John Wiley, New York, 2001
An excellent textbook, covering in depth most of the lectures. Free download on several servers.
[2] Gonzales R. C., Woods R. E., Digital Image Processing (2nd ed.), Prentice Hall, 2002
A good aid for exercises.
Supplementary text to Lecture 5.
MATERIALS FOR LABS
Will be published during the semester.
LABS
We are going to take a step into the world of Image Processing using python. In order to do that, you should know the basics of Python. Please go through the following section before attending labs. Necessary information for completing all labs will be provided here. If you have any questions please contact novozamsky@utia.cas.cz
The exercises will be in Czech, but all materials and texts will be in English.
REQUIREMENTS
Zoom account - the webcam will be on !
GitHub account
Gmail account for Colab is recommended
Browser
PYTHON CRASH COURSE
LABS START HERE!
At the beginning of every lab session, join Zoom Meeting, where you receive a link for a GitHub classroom assignment. It contains all codes for online/offline studying.
Zoom link: https://cuni-cz.zoom.us/j/91367957348
Get a link from your teacher, e.g. https://classroom.github.com/a/YspbK****
Sign in to GitHub
If necessary, click on the "Authorize GitHub Classroom" button
Find your name in the list (only for the first assignment)
Click on "Accept the assignment"
Click on "Authorize Google Colab"
Click on "GitHub" (or File > Open Notebook > GitHub)
Type your GitHub user name (check "Include private repos")
Find the desired repository and the notebook file
Loading (In case of "Error", press "Retry")
If necessary, click on the "Authorize with GitHub" button
Edit the notebook, finish excercises
Save in GitHub (File > Save a copy in GitHub)