News

• December 22, 2017: ninety-minute lecture.
• December 1, 2017: Exercise instead of lecture.
• October 25, 2017: Exercise instead of lecture.
• October 20, 2017: The lecture has to end at 14:00 due to a subsequent event.
• October 18, 2017: The lecture starts.

Exam

The exam will be a 90 min written exam. You will not be able to take any written material into the exam, but a formula sheet will be provided. You should take a calculator and a pen (not a pencil).

• The exam is on 26.02.2018, 08:00-09:30
• The repetition exam is on 04.04.2018, 15:30 - 17.00

Description

The course gives an overview of application areas and techniques in Artificial Intelligence. The course introduces the principles and techniques of Artificial Intelligence based on the textbook of Russell and Norvig (see below). The course covers the following topics:

• Task environments and the structure of intelligent agents.
• Solving problems by searching: breadth-first search, uniform-cost search, depth-first search, depth-limited search, iterative deepening search, greedy best-first search, A* search.
• Constraint satisfaction problems: defining constraint satisfaction problems, backtracking search for constraint satisfaction problems, heuristics for backtracking search, interleaving search and inference, the structure of constraint satisfaction problems.
• Logical agents: propositional logic, propositional theorem proving, syntax and semantics of first-order logic, using first-order logic, knowledge engineering in first-order logic, reducing first-order inference to propositional inference, unification and lifting, forward chaining, backward chaining, resolution.
• Bayesian networks: acting under uncertainty, basics of probability theory, Bayesian networks, inference in Bayesian networks, approximate inference in Bayesian networks.
• Hidden Markov models: time and uncertainty, inference in hidden Markov models (filtering, prediction, smoothing, most likely explanation), approximate inference in hidden Markov models.
• Rational decisions: introduction to utility theory, utility functions, decision networks, the value of information, Markov decision processes, value iteration, policy iteration, partially observable Markov decision processes.
• Learning: types of learning, supervised learning, learning decision trees.
• Introduction to robotics: robot hardware, robotic perception, path planning, planning uncertain movements, control of movements, robotic software architectures, application domains.

Material

• The material is provided through the moodle website.
• Last year's moodle website (for long-term preview) is here.

Literature

• Stuart Russell and Peter Norvig, Artificial Intelligence: A Modern Approach (3rd Edition). Prentice Hall, 2009. ISBN 0-13-604259-7.
• German edition: Russel/Norvig: Künstliche Intelligenz: Ein moderner Ansatz, 3. Auflage, Pearson, 2012