Intelligent Agents and Automated Reasoning

Uitgebreide vaknaam	Intelligent Agents and Automated Reasoning
Leerdoelen	After this course, the students should be able to: 1) explain the concepts and principles of artificial intelligence, the concepts of an intelligent agent and rational decisions; 2) name and explain sub-areas of intelligent reasoning, and types of problems that can be solved by approaches in these sub-areas, including: search, logic, constraint satisfaction problem, knowledge representation, game theory, AI Planning, reasoning under uncertainty, real-time actuation; 3) formally define a reasoning task, solve it, and prove the rationality of their solution; 4) program an automated intelligent agent that can perform rational decision making.
Omschrijving	The course provides an overview of how machines (robots, automated reasoning systems) think and make decisions and goes over several reasoning domains, algorithms, and approaches (some related to each other, some unrelated), with progressive difficulty. Topics that will be discussed on the course, include: search, logic (propositional, predicate, first-order), constraint satisfaction problem, knowledge representation, game theory, adversarial agents, AI Planning (classical and hierarchical task network), reasoning under uncertainty (bayesian, hidden markov models), real-time actuation, stateless vs stateful reasoning. Central to the course is the concept of rational decision making – a logical approach to choosing an action out of several alternatives, based on the available information and maximization of the desired outcome.
Uren per week
Onderwijsvorm	~computerpracticum, Hoorcollege (LC) (Computer Labs, Lectures)
Toetsvorm	Opdracht (AST), Schriftelijk tentamen (WE) (Assignments (A), both theoretical and practical in nature, will be worth 60% of the total grade. The final written exam (E) will be worth 40% of the total grade. In order to pass, both (A) and (E) grades must be at least 5.0, and the total weighted final grade must be at least 6.0, when rounded to the nearest half-point.)
Vaksoort	bachelor
Coördinator	V. Degeler, PhD.
Docent(en)	V. Degeler, PhD. , D. Düstegör, PhD.
Verplichte literatuur	Titel Auteur ISBN Prijs Artificial Intelligence: A Modern Approach, 3 rd edition. (mandatory) Stuart Russell, Peter Norvig
Entreevoorwaarden	The course unit assumes background experience in basic programming, data structures and algorithms, discrete mathematics, logic and sets.
Opmerkingen	Other editions of the book are acceptable as well, although reading material will be given according to the 3 rd edition. --- All CS bachelor courses have limited enrollment: - CS students can always enter each course, regardless of whether the course is mandatory for them or not. - A maximum of only 20 places per course is available for non-CS students. These places are filled on a first-come-first-served basis, with priority given to students with a strong CS-related background (e.g., CS exchange students, AI students, etc.). These students need to meet the course prerequisite requirements as mentioned on Ocasys. - Six weeks before the course starts, the 20 students that can join are selected and added to the course. If you enroll after this date, you will be placed on the waiting list. For more info about the enrollment procedure, see https://student.portal.rug.nl/infonet/studenten/fse/programmes/bsc-cs/general/vakintekening-procedure#cap
Opgenomen in	Opleiding Jaar Periode Type BSc Computing Science  (Optional electives for CS students) 3 semester I b keuze