26E113TMU - Machine learning techniques

Course specification
Course title		Machine learning techniques
Acronym		26E113TMU
Study programme		Electrical Engineering and Computing
Module		Computer Engineering and Informatics
Type of study		bachelor academic studies
Lecturer (for classes)		professor PhD Boško Nikolić asst. professor PhD Vladimir Jocović
Lecturer/Associate (for practice)		asst. MA Adrian Milaković, electrical and computer engineer
Lecturer/Associate (for OTC)		asst. MA Adrian Milaković, electrical and computer engineer
ESPB		6.0	Status	elective
Condition		Intelligent systems
The goal		Introducing students to the basic concepts and techniques of machine learning. During the course, students will study the most popular models for designing, implementing, and testing these types of applications (natural language processing, computer vision, speech recognition, pattern recognition).
The outcome		Students will gain knowledge of advanced machine learning techniques and methodologies. The course covers neural networks, including convolutional and recurrent models, and hyperparameter tuning. Its main outcome is equipping students with skills and tools to solve complex AI and ML problems in real-world scenarios.
Contents
URL to the subject page		http://ri.etf.bg.ac.rs/
Contents of lectures		The course covers fundamentals, training, and design of neural networks (regularization, optimization, hyperparameters). It explores CNNs, RNNs, and transformers, with applications in classification, detection, NLP, and speech recognition. Students will use TensorFlow, PyTorch, and Keras, and analyze examples from industry and academia.
Contents of exercises		Analyzing and solving practical tasks and demonstrating how to overcome certain problems with machine learning and deep learning techniques. Practical work with Tensorflow and Keras tools on examples of different datasets for text, image, and pattern recognition analysis
Literature
Deep Learning, Ian Goodfellow, Yoshua Bengio, Aaron Courville, The MIT Press, 2017 (Original title) Neural Networks and Deep Learning, Michael Nielsen, Determination Press, 2019 (Original title) Deep Learning with Python, 2nd Edition, Francois Chollet, Manning, 2021 (Original title) Deep Learning for Natural Language Processing, Stephan Raaijmakers , Manning 2022 (Original title)
Number of hours per week during the semester/trimester/year
Lectures	Exercises	OTC	Study and Research	Other classes
2	2	1
Methods of teaching		Analyzing and solving practical tasks and demonstrating how to overcome certain problems with machine learning and deep learning techniques. Practical work with Tensorflow and Keras tools on examples of different datasets for text, image, and pattern recognition analysis
Knowledge score (maximum points 100)
Pre obligations		Points	Final exam	Points
Activites during lectures			Test paper	30
Practical lessons			Oral examination
Projects		40
Colloquia		30
Seminars