To teach the mathematical approaches that form the basis and are useful in design, learning, or evaluation processes of artificial intelligence methods, and in this context to carry out thorough examinations of signal analysis, probability theorem, optimization techniques, decision processes, statistical tests, information theory, fuzzy logic, belief theory, decision processes, and deep networks.
20
Contribution of the Course to Professional Development
To have students comprehend the operational logic behind commonly used artificial intelligence methods and build the necessary mathematical foundations to design similar intelligent methods.
21
Learning Outcomes:
1
Knowing the requisites and deficiencies of the artificial intelligence systems;
2
Being able to analyze real problems so as to solve them with artificial intelligence systems;
3
Being able to associate probability, information, and belief theories with representation, learning, or testing phases of the intelligent systems;
4
Being knowledgeable about optimization techniques that enable learning;
5
Being able to describe the differences between the fuzzy logic-based and probability-based applications;
6
Being able to learn empirical or statistical models with decision trees, artificial neural networks, or deep networks;
7
Being able to ground the evolution of learning strategies from artificial neural networks to deep networks;
22
Course Content:
Week
Theoretical
Practical
1
Limits of artificial intelligence, human-level artificial intelligence, expert systems
2
Signal terminology and signal characteristics
3
Probability theory, conditional probability, Bayes’ theorem, independence
4
Probability-based classifiers, data preparation, evaluation
5
Basics of learning and optimization techniques
6
Optimization in artificial neural networks, backpropagation, network parameters
7
Random processes and decision making
8
Statistical tests, information theory
9
Fuzziness and belief theory
10
Applications of fuzzy logic
11
Hopfield networks, Boltzmann machines
12
Deep restricted Boltzmann machines
13
Variational autoencoders
14
Reinforcement learning, Markov decision processes
23
Textbooks, References and/or Other Materials:
1. Jackson, P. C., 2019. Toward Human-Level Artificial Intelligence, Dover Publications Inc, ISBN-10: 0486833003 ISBN-13: 978-0486833002. 2. Bender, E. A., 1996. Mathematical Methods Artificial Intelligence, Wiley-IEEE Computer Society Pr, ISBN-10: 0818672005 ISBN-13: 978-0-818-67200-2. 3. Stone, J. V., 2019. Artificial Intelligence Engines: A Tutorial Introduction to the Mathematics of Deep Learning, Sebtel Press, ISBN-10: 0956372813 ISBN-13: 978-0956372819.
24
Assesment
TERM LEARNING ACTIVITIES
NUMBER
PERCENT
Midterm Exam
1
20
Quiz
0
0
Homeworks, Performances
3
30
Final Exam
1
50
Total
5
100
Contribution of Term (Year) Learning Activities to Success Grade
50
Contribution of Final Exam to Success Grade
50
Total
100
Measurement and Evaluation Techniques Used in the Course
written exams, assignments, research report, presentation
Information
25
ECTS / WORK LOAD TABLE
Activites
NUMBER
TIME [Hour]
Total WorkLoad [Hour]
Theoretical
14
3
42
Practicals/Labs
0
0
0
Self Study and Preparation
14
3
42
Homeworks, Performances
3
18
54
Projects
0
0
0
Field Studies
0
0
0
Midtermexams
1
16
16
Others
0
0
0
Final Exams
1
24
24
Total WorkLoad
178
Total workload/ 30 hr
5,93
ECTS Credit of the Course
6
26
CONTRIBUTION OF LEARNING OUTCOMES TO PROGRAMME QUALIFICATIONS