Admission Requirements
Knowledge of probability and optimization theory, basic analysis of algorithms, and programming skills

Learning Outcomes

Following the successful completion of the course the following main learning outcomes are expected:

Knowledge:

• Name widely used AI / ML methods and what specific problems they can be used to solve.
• Explain the possibilities and principles of the AI / ML algorithm application.
• Discuss decision-making systems, basic components and design principles.
• Explain the steps of the widely used AI / ML algorithms.
• Describe the weak points (things to consider) for the most popular AI / ML algorithms.

Skills:

• Implement machine learning methods in MATLAB or Python environment.
• Make a comparative analysis of the results obtained using different models.
• Design an AI system for solving a practical problem.
• Critically assess the quality of the AI system.

Competencies:

• Select and apply the most suitable artificial intelligence (AI) /machine learning (ML) method(s) depending on the given problem, and argue the decision in the discussion.
• Work independently, systematically, and responsibly, including time planning and choosing effective ways of working, taking the initiative, and assuming personal responsibility.
• Integrate AI / ML methods into the decision-making systems.
• Critically review the existing AI solutions and systems.

Programme
Computer Intelligent paradigms and Decision-making theory. Optimization and Search: Evolutionary learning; Gradient optimization methods. Supervised learning: Decision Trees; K nearest neighbours; Probability-based learning; The multi-Layer perceptron and radial basis functions; Convolutional Neural Network. Unsupervised learning: K-means method; Fuzzy C-means method and fuzzy logic; Self-organizing map. Reinforcement learning: Markov decision process; Q-learning; Monte-Carlo tree search

References

• Rebala, Gopinath. “An introduction to machine learning”, 2019
• Mohammed, Mohssen, Muhammad Badruddin Khan, and Eihab Bashier Mohammed Bashier “Machine learning: algorithms and applications”, 2017
• Andrea Lonza. Reinforcement Learning Algorithms with Python: Learn, understand, and develop smart algorithms for addressing AI challenges, 2019

Teaching Methodology
Challenge-based learning, laboratory classes, lectures, tutorials

ECTS Credits
7.5 ECTS

Period
II semester (STEM)

Examination methodology
Oral examination, problem-solving tasks, presentations. The ten-grade scale and the cumulative evaluation system are applied. The module’s final evaluation consists of the sum of multiplications of the grades of the intermediate assessments and the final assessment multiplied by weighting coefficients (percentage components).

Relevance
Nowadays artificial intelligence (AI) is met almost in every step of life. It is important to understand the possibilities and limitations of AI to make the most of it. During the course, students work in groups to identify and solve a particular problem in the given field with respect to various aspects (social, economic, technical, ethical, etc.). They develop knowledge in the field of artificial intelligence (AI) by analyzing distinct AI methods in the experimental environment and applying them in a close-to-real-life environment in which separate components of data preparation, AI application and interpretation of the results should be treated as an integral system with appropriate links. The analysis of the system, its parts, and the relationship between them enables us to identify the problem and suggest a sustainable solution with consideration of different perspectives by various society members.

Release
Existing course but revised