This seminar is designed for students with prior knowledge of machine learning who wish to explore and apply ML/DL techniques to real-world challenges. The course follows a project-based approach, where the final project is developed throughout the semester, beginning in week one. Students will move from problem definition and literature review to data acquisition, model training, and evaluation. Each week, the lecturer introduces relevant concepts and methods aligned with the project stage, while teams present their ongoing progress. The seminar culminates in final project presentations and written reports.
This course provides an introduction to the foundations and applications of Generative Artificial Intelligence (GenAI). Building on prior knowledge in machine learning and in parallel with deep learning studies, students will explore the theory and practice of generative models, including Transformers, GANs, Variational Autoencoders, Normalizing Flows, and Diffusion Models. The course combines instructor-led lectures with student-led presentations and a semester-long team-based research project. Students will critically engage with recent scientific literature, design and implement generative models on appropriate datasets, and present their findings in both oral and written form. The course emphasizes not only technical skills but also teamwork, critical thinking, and scientific communication.
During the course, methods for image processing and computer vision will be presented, with an emphasis on techniques currently used in academia and industry. Topics will include basic image processing, feature detection of objects in images, image quality enhancement, object tracking and classification, color identification, depth, lighting, and shading, scene understanding in images, and more.
This course introduces students to the fundamental principles of machine learning, focusing on understanding and applying various algorithmic models. Students will learn how to analyze and interpret complex datasets, build regression and classification models, and apply unsupervised learning techniques. Throughout the course, students will be assigned homework exercises to practice the concepts learned, and a final exam will be held
