BE5B33KUI: Cybernetics and Artificial Intelligence

Lab instructors: Pavel Šindler, Petr Pošík

The course introduces the students to the field of artificial intelligence and gives the necessary basis for designing machine control algorithms. It advances the knowledge of state space search algorithms by including uncertainty in state transitions. Students are introduced into reinforcement learning for solving problems when the state transitions are unknown. Bayesian decision task introduces supervised learning. Learning from data is demonstrated on a linear classifier. Students practice the algorithms in computer labs.

Disclaimer: This course is NOT about deep neural nets, LLMs, chatbots, etc., but the last part of the course about machine learning builds some foundation for these topics.

Prerequisities: The course expects basic knowledge of probability and linear algebra. We expect that students are able to write decent computer programs in a higher level language and have basic knowledge of data structures. Python will be used in computer labs.

The course runs in a hybrid regime. The lectures and lab sessions for online participants are streamed and recorded via BigBlueButton in BRUTE. Students enrolled in EECS or other regular presence study programs at CTU are expected to physically attend lectures/labs.

lectures | labs | literature | schedule 2025/2026

Grading

The course uses the usual grading scheme:

Points	[0, 50)	[50, 60)	[60, 70)	[70, 80)	[80, 90)	[90, 100+]
Grade	F	E	D	C	B	A

The score is composed of three main components: work during semester, midterm exam, and final exam. (There will be some bonus points for discussions/quizzes during the computer labs.)

Component	Points	Required
Work during semester	45	18 + all 5 mandatory assignements accepted
Midterm exam	15	0 (you can still successfully finish the course even if you fail in midterm)
Final exam	40	18

At least 18 points for the homework (lab assignments) are needed before going to the final exam.
In addition, you need to submit all of the 5 bigger assignments (Search, Reversi, SDP, RL, classification/recognition) at least at a minimum level of functionality (at least 1 point from the auto-evaluations, ignoring eventual minus points for missing the deadline).