Instructor: Sungwoong Kim
Time: Tue 12:00 - 14:45 (break time: 13:15 - 13:30)
Room: Jung Woonoh IT & General Education Center, 519
Contact: swkim01@korea.ac.kr or LMS
This course covers the foundations and recent advances of deep learning. We begin with basic architectures, including MLPs, CNNs, and RNNs, to understand how deep learning has evolved from traditional machine learning and classical neural networks. We then explore advanced topics such as Bayesian Deep Learning, Neural Processes, Representation Learning, and Foundation Modeling. The second half focuses on generative modeling, which is central to recent deep learning and AI development, exploring how algorithms and underlying technologies of different generative models are related. Finally, we cover deep reinforcement learning in the development of modern AI agents.
Throughout the course, students will learn the backgrounds and key factors of deep neural networks and deep learning algorithms. In particular, the course will cover basic deep learning and deep neural network architectures as well as recent models and algorithms including deep generative models, foundation models, and deep reinforcement learning. Eventually, the course aims students to have enough knowledge and moreover an insight for the corresponding developments and researches.
Basic knowledge of probability and machine learning is strictly required. Students without prior knowledge or coursework in these areas may find it difficult to keep up with the course.
Lecture notes will be the main material of the course, and these do not come from a single textbook.
There will be no specific assignment. The evaluation will be based on the participation, attendance, midterm exam, and final exam. Active participation is highly encouraged, being reflected to the evaluation.
Participation (20%)
Attendance (20%)
Midterm Exam (30%)
Final Exam (30%)