Artificial Intelligence

Course Description

There are many cognitive tasks that are simple for the human, but very difficult for the computer. Artificial intelligence (AI) tries to solve these kind of problems. Traditional approach to AI is based on symbolic knowledge representation and reasoning as symbol manipulation. Biologically inspired connectivistic approach simulates cognitive abilities of the brain. The aim of the course is to introduce different AI approaches and to give an overview AI methods, including methods for knowledge representation, automatic reasoning, problem solving, learning, and optimization.

General Competencies

Students will have an overview of the different approaches to AI and different AI methods. They will understand the drawbacks and the advantages of different approaches and will be able to recognize the types of problems that can be tackled successfully using AI methods. Students will gain practical programming experience in solving diverse AI problems, including state space search, game playing, automatic inference, logic programming, neural networks, and biologically inspired optimization.

Learning Outcomes

  1. define the basic concepts of artificial intelligence
  2. distinguish between symbolic and connectivistic approaches to AI
  3. apply state search algorithms and biologically inspired optimization algorithms on basic problems
  4. solve basic problems using logic programming
  5. apply inference algorithms on basic logical problems
  6. compare among various approaches to representing uncertainty
  7. assess the applicability of different AI methods on a given AI problem
  8. review the philosophical aspects of artificial intelligence

Forms of Teaching

Lectures

Lectures are given for 13 weeks (3 hours each). Lectures are divided in two study periods. There will be a midterm exam after the first period and a final exam after the second period.

Exams

Midterm exam and final exam; short quizzes during the semester.

Laboratory Work

There will be 4-5 lab take-home lab assignments, demonstrated to the instructor or the lab assistent.

Consultations

Weekly office hours.

Grading Method

Continuous Assessment Exam
Type Threshold Percent of Grade Comment: Percent of Grade
Laboratory Exercises 25 % 25 % 25 % 25 %
Quizzes 0 % 5 % 0 % 0 %
Mid Term Exam: Written 0 % 35 % 0 %
Final Exam: Written 0 % 35 %
Exam: Written 50 % 37.5 %
Exam: Oral 37.5 %
Comment:

Exams after the final exam are written and oral, together contributing 75% to the final grade. Students are required to score above 50% on written exam to be admitted to the oral exam.

Week by Week Schedule

  1. Overview of the artificial intelligence. History of AI. AI research areas and newest trends. Relation to other disciplines. Intelligence and the Turing test.
  2. Solving problems by searching state space search. Blind search strategies.
  3. Informed search strategies. A* algorithm. Constraint satisfaction problems. Game playing. Min-max algorithm.
  4. Knowledge and reasoning. First order logic. Theorem proving. Unification. Resolution rule.
  5. Logic programming. Prolog.
  6. Semantic networks, frames and rules. Ontologies. Expert systems.
  7. Natural language processing.
  8. Midterm exam.
  9. Uncertain knowledge and reasoning. Probability-based knowledge representation. Bayes rule. Fuzzy logic and fuzzy inference.
  10. Introduction to machine learning. Naive Bayes classifer.
  11. Connectivistic approaches to AI. Artificial neural networks. Perceptron algorithm. Backpropagation algorithm.
  12. Computational intelligence. Genetic algorithm. Ant colony optimization.
  13. Embodied AI. Behavior-oriented AI.
  14. Philosophical foundations of AI. Wrap up.
  15. Final exam.

Study Programmes

University undergraduate
Computer Science (module)
(6. semester)
Software Engineering and Information Systems (module)
Elective Courses (6. semester)

Literature

Stuart Russell and Peter Norvig (2009.), Artificial Intelligence: A Modern Approach, Prentice Hall, London
Rolf Pfeifer and Christian Scheier (2001.), Understanding Intelligence, The MIT Press

Lecturers

Laboratory exercises