Machine learning is the study of algorithms and statistical methods that computers use to "learn" patterns and inferences in order to perform a specific task without explicit instruction. This course mainly aims at providing mathematical/statistical methods, which are essential in machine learning. A wide range of topics will be covered, including but not limited to density estimation, latent variable models, mixture models, clustering, classification, dimensionality reduction, regression, support vector machines, kernel methods, multi-layer perceptrons, and deep learning.

Basic understanding of probability, statistics, and linear algebra

Assignments: 40% (five or six assignments)
Midterm exam: 30%
Final exam: 30%

(zero tolerance for plagiarism or cheating)

Machine Learning: A Probabilistic Perspective (by K. Murphy, 2012)
Pattern Recognition and Machine Learning (by C. Bishop, 2006, Springer)

(those are publicly available online)

(Tentative schedule: https://docs.google.com/spreadsheets/d/1hyLHHOfrCPbpAXmP0PT6z1V5xQ7DvIHlFsf_4cy3sUg/edit?usp=sharing)
1-1 2025/2/18 Introduction to Machine Learning
1-2 2025/2/20 Review on Probability Theory
2-1 2025/2/25 Generative Model and Parameter Estimation (1)
2-2 2025/2/27 Generative Model and Parameter Estimation (2)
3-1 2025/3/4 Linear Regression
3-2 2025/3/6 Linear Regression
4-1 2025/3/11 Logistic Regression: Classification
4-2 2025/3/13 Logistic Regression: Classification
5-1 2025/3/18 Support Vector Machine and Constraint Optimization
5-2 2025/3/20 Multi-class Classification
6-1 2025/3/25 Neural Network
6-2 2025/3/27 Neural Network
7-1 2025/4/1 Some Insights on Model Selection
7-2 2025/4/3 Some Insights on Model Selection
8-1 2025/4/8 Midterm Exam (2 hrs, 11:00-13:00)
8-2 2025/4/10 -
9-1 2025/4/15 Graphical Model
9-2 2025/4/17 Belief Propagation (BP)
10-1 2025/4/22 Graph Construction
10-2 2025/4/24 Clustering
11-1 2025/4/29 EM algorithm
11-2 2025/5/1 Principal Components Analysis
12-1 2025/5/6 -
12-2 2025/5/8 Non-linear PCA and Autoencoder
13-1 2025/5/13 VAE (1)
13-2 2025/5/15 VAE (2)
14-1 2025/5/20 GAN
14-2 2025/5/22 MDP and Dynamic Programming
15-1 2025/5/27 RL(1)
15-2 2025/5/29 RL(2)
16-1 2025/6/3 Final Exam (2 hrs, 11:00-13:00)
16-2 2025/6/5 -

Assignment: every two or three weeks
Exam: midterm and final (closed book)

All lectures, assignments, and exams will be given in English. However, students may use Korean if preferred.