Cybernetics and Artificial Intelligence
| Code | Completion | Credits | Range | Language |
|---|---|---|---|---|
| B3B33KUI | Z,ZK | 6 | 2P+2C | Czech |
- Garant předmětu:
- Tomáš Svoboda
- Lecturer:
- Petr Pošík, Tomáš Svoboda
- Tutor:
- Jana Kostlivá, Petr Pošík, Tomáš Svoboda, Pavel Šindler
- Supervisor:
- Department of Cybernetics
- Synopsis:
-
The course introduces the students into 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 transition. Students are introduced into reinforcement learning for solving problems when the state transitions are unknown, which also connects the artificial intelligence and cybernetics fields. Bayesian decision task introduces supervised learning. Learning from data is demonstrated on a linear classifier. Students practice the algoritms in computer labs.
- Requirements:
-
Basic knowledge of linear algebra and programming is assumed. Experience in Python and basics of probability is an advantage.
- Syllabus of lectures:
-
What is artificial intelligence and what cybernetics.
Solving problems by search. State space.
Informed search, heuristics.
Games, adversarial search.
Making sequential decisions, Markov decision process.
Reinforcement learning.
Bayesian decision task.
Learning from examples. Linear classifier. Nearest neighbors method.
Empirical evaluation of classifiers ROC curves.
- Syllabus of tutorials:
-
During computer labs and at home, students will implement several algorithms introduced at lectures. The emphasis will be put to testing the functionality of their implementation. When exercising classification problems, we will also discuss the topics training and testing data, crossvalidation and ROC curve. A technical report will be required for some of the tasks.
- Study Objective:
-
The course introduces the students into 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 transition. Students are introduced into reinforcement learning for solving problems when the state transitions are unknown, which also connects the artificial intelligence and cybernetics fields. Bayesian decision task introduces supervised learning. Learning from data is demonstrated on a linear classifier. Students practice the algoritms in computer labs.
- Study materials:
-
Stuart J. Russel and Peter Norvig. Artificial Intelligence, a Modern Approach, 3rd edition, 2010
Richard O. Duda, Peter E. Hart, David G. Stork. Pattern Classification, 2nd edition. 2000
Christopher M. Bishop. Pattern Recognition and Machine Learning. 2006
- Note:
- Further information:
- http://cw.fel.cvut.cz/wiki/courses/b3b33kui/start
- Time-table for winter semester 2023/2024:
- Time-table is not available yet
- Time-table for summer semester 2023/2024:
-
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 Tue Wed Thu Fri - The course is a part of the following study plans:
-
- Cybernetics and Robotics 2016 (compulsory course in the program)
- Medical electronics and bioinformatics (compulsory elective course)
- Cybernetics and Robotics 2016 (compulsory course in the program)
During a review of study plans, the course A3B33KUI can be substituted for the course B3B33KUI.
In order to register for the course B3B33KUI, the student must have registered for the required number of courses in the group BEZBM no later than in the same semester.
The requirement for course B3B33KUI can be fulfilled by substitution with the course BE5B33KUI.