Intelligent Systems

Course code

IFI6057.DT

old course code

IFI6057

Course title in Estonian

Intelligentsed süsteemid

Course title in English

Intelligent Systems

ECTS credits

4.0

approximate amount of contact lessons

56

Teaching semester

autumn

Assessment form

Examination

lecturer of 2019/2020 Autumn semester

Jaagup Kippar (eesti keel) e-toega kursus

lecturer of 2019/2020 Spring semester

lecturer not assigned

Course aims

The course gives practical knowledge about the algorithms used in the field of artificial intelligence and the skills to deploy them.

Brief description of the course

Introduction to the principles and algorithms used in the field of artificial intelligence. Solving problems by searching. Heuristics. Fundamentals of logical and probabilistic reasoning. Bayes' rule. Machine learning: supervised learning, decision tress, classifying with linear models and reinforcement learning. Neural networks. Ethics of artificial intelligence.

Independent work

Homework (programming and individual study)

Learning outcomes in the course

The student:

- is capable of formulating artificial intelligence problems as state space search

- knows tree search and local search algorithms and can apply them; including BFS, DFS, A* and hill climbing.

- Can describe the behaviour and parameters of tree search and local search

- is familiar with modern approaches to combinatorial search (metaheuristics)

- is knowledgeable about the principles of logical and probabilistic reasoning

- can solve problems involving propositional logic and Bayesian probability

- is familiar with some problems in machine learning (classification, learning behaviour) and approach methods (decision tree, linear classifiers, neural networks, reinforcement learning)

- can select and apply a suitable machine learning method

- can use a modern machine learning package to solve machine learning tasks (scikit-learn or weka).

- is capable of formulating artificial intelligence problems as state space search

- knows tree search and local search algorithms and can apply them; including BFS, DFS, A* and hill climbing.

- Can describe the behaviour and parameters of tree search and local search

- is familiar with modern approaches to combinatorial search (metaheuristics)

- is knowledgeable about the principles of logical and probabilistic reasoning

- can solve problems involving propositional logic and Bayesian probability

- is familiar with some problems in machine learning (classification, learning behaviour) and approach methods (decision tree, linear classifiers, neural networks, reinforcement learning)

- can select and apply a suitable machine learning method

- can use a modern machine learning package to solve machine learning tasks (scikit-learn or weka).

Assessment methods

Grading is done using points system.

Participation and solving problems in labs: up to 25 points.

Homework assignment: up to 15 points.

Written examconsisting of problems and questions (variable weight): up to 70 points.

Final grade:

0-49 points- F

50-59 points- E

60-69 points- D

70-79 points- C

80-89 points- B

90 or more points- A

Participation and solving problems in labs: up to 25 points.

Homework assignment: up to 15 points.

Written examconsisting of problems and questions (variable weight): up to 70 points.

Final grade:

0-49 points- F

50-59 points- E

60-69 points- D

70-79 points- C

80-89 points- B

90 or more points- A

Teacher

Jaagup Kippar

Prerequisite course 1

Replacement literature

M. Koit, T. Roosmaa. Tehisintellekt. Tartu, TÜ Kirjastus, 2011. (http://dspace.utlib.ee/dspace/handle/10062/28296)

Russell, S.J and Norvig, P. Artificial intelligence: a modern approach, third edition, Prentice Hall. 2009.

Algorithms and Architectures of Artificial Intelligence, IOS Press, 2007.

Russell, S.J and Norvig, P. Artificial intelligence: a modern approach, third edition, Prentice Hall. 2009.

Algorithms and Architectures of Artificial Intelligence, IOS Press, 2007.