Computational Game Theory (BE4M36MAS) Winter 2021/2022
The course provides an introduction to concepts, models and algorithms for autonomous agents and multi-agent systems. Game theory is the key formalism used in multi-agent systems that describes and defines optimal behavior of an agent while explicitly reasoning about plans and goals of other agents. In the course, we will explains key multiagent models and algorithms, both for cooperative and non-cooperative settings. Upon successful completion of the course, students will be able to understand main multi-agent concepts, be able to map real-world multi-agent problems to multiagent formal models and apply algorithmic techniques to solve them.
Lectures:
Tutorials:
Tuesday 11:00-12:30, 14:30-16:00, 16:15-17:45, 21.9.2021 – 4.1.2022, KN:E-307
Main contacts:
Regarding lectures/course: branislav.bosansky@agents.fel.cvut.cz
Regarding tutorials: dominik.seitz@aic.fel.cvut.cz (main contact, EN only), michal.sustr@aic.fel.cvut.cz (backup contact)
Links
Grading
Both the course assessment and exam are required to pass the course. The final grade (A..F) will be determined by the sum of points obtained from the assessment and exam (<50 = F, 50-59 pts = E, …, 90-100 pts = A).
Assessment
Minimum of 25 pts is required from the labs
Game theory: max grading: 14 pts.
Coalitional game theory: max grading: 12 pts.
Midterm Test: max grading 24 pts.
The penalty for submitting the homework assignment after the deadline, but no later than 24 hours after the deadline, is 20% of the points.
The penalty for submitting the homework assignment later than 24 hours after the deadline is 100% of the points.
Exam
Minimum of 25 pts is required from the exam (out of maximum 50 pts).
The exam is written. In selected cases, a brief oral part to clarify answers follows.
The form of exam/requirements can change depending on the current COVID restrictions.
Exam topics correspond to the topics covered by lecture slides
Course assessment is required prior to attending an exam
Update January 2022, Form of the exam
The exam is written, you have 150 minutes for the exam
The exam will be graded and the points will be put into the BRUTE system – we will have online calls with those who wish to consult their results of the exam, discuss correct answers, etc. upon request
To minimize the probability of COVID transmission, please:
use FFP2 face mask and keep it on all the time
disinfect your hands upon entering the building
keep your distance from other colleagues
if possible, get yourself tested before the exam – if you have a positive test or if you are not feeling well, please, do not come to the exam and sign-up to the next exam date. If needed, we will open additional exam dates.
Dates:
12.1. 13:30 – 16:30 (KN:E-107)
26.1. 9:00 – 12:00 (KN:E-107)
2.2. 9:00 – 12:00 (KN:E-107)
Exam from the last years: example
Lectures
(subject to permutation)
For accessing the videos, please, log-in using your Google FEL account.
Tutorials
Reading Resources
[Shoham] Shoham, Y. and Leyton-Brown, K.: Multiagent Systems: Algorithmic, Game-Theoretic, and Logical Foundations, Cambridge University Press, 2008, ISBN 9780521899437.
-
[AIMA] Russel, S. a Norvig, P.: Artificial Intelligence: A Modern Approach (2nd edition), Prentice Hall, 2003
[Wooldridge] Wooldridge, M.: An Introduction to MultiAgent Systems, John Wiley & Sons Ltd, 2002, ISBN 0-471-49691-X.
Tutorial Resources
For running IntelliJ Idea on local machines use command /opt/idea-IC-173.4548.28/bin/idea.sh.