COMP 5212: Machine Learning [Fall 2023]
Wednesday, Friday 16:30-17:50 @ Room 6591 (Lift 31-32)
Overview
- Instructor: Minhao Cheng (minhaocheng@cse.ust.hk)
- Office hours: Tuesday 13:00-14:30 @ CYT 3004
- Teaching assistants:
- Zeyu Qin’s (zeyu.qin@connect.ust.hk) office hours: Thursday 10:00 - 11:00 @ Room 1008
- Sen Li’s (slien@connect.ust.hk) office hours: Monday 14:00-15:00 @ LG4 RPG Hub 4002
- Canvas: COMP5212
Announcements
[8 Sep 2023] Today’s lecture will be online over ZOOM due to the heavy rain.
[1 Sep 2023] Welcome to COMP5212!
Description
In this course, we will cover some classical and advanced algorithms in machine learning. Topics include: Linear models (linear/logistic regression, support vector machines), Non-linear models (tree-based methods, kernel methods, neural networks), learning theory (hypothesis space, bias/variance tradeoffs, VC dimensions). The course will also discuss some advanced topics of machine learning such as testing-time integrity in trustworthy machine learning and neural architecture search in AutoML.
Prerequisites
Basic knowledge in numerical linear algebra, probability, and calculus.
Grading Policy
- Homework (40%)
- 3 Written homeworks
- 2 Programming homeworks
- Term project (35%)
- Final exam (25%)
Late submission policy:
Late submissions are accepted up to 2 days after the due date, with 10% (of the total grade of the item) penalty per day.
Term projects
Students will work on a open-topic research project with groups. Each group could only be consisted with less or equal than 4 members (<=4). Feel free to discuss with me offline for the topic choice.
Tentative Schedule and Material
Date | Topic | Slides | Readings&links | Assignments |
---|---|---|---|---|
Wed 6/9 | Overview of Machine Learning | lecture_0 | ||
Fri 8/9 | Math Basics | lecture_1 | Matrix Calculus:Derivation and Simple Application HU Pili, DL Chapter 2.1 & 2.2 &2.3 | |
Wed 13/9 | Linear models | lecture_2 | ||
Fri 15/9 | Optimization | lecture_3 | Convex Optimization Boyd and Vandenberghe Chapter 3.1, Numerical Optimization Nocedal and Wright Chapter 3.1 | |
Wed 20/9 | Stochastic gradient descent and its variants | lecture_4 | Written_hw1 out | |
Fri 22/9 | Support Vector Machine, Polynomail nonlinear mapping, Kernel method, | lecture_5 | Stanford CS 229 notes | |
Wed 27/9 | Polynomail nonlinear mapping, Kernel method | lecture_6 | Stanford CS 229 notes | |
Fri 29/9 | Learning theory | lecture_7 | Symmetrization | |
Wed 4/10 | Uniform convergence, growth function | lecture_8 | Bias/variance tradef off | Programming_HW1 out |
Fri 6/10 | VC Dimension | lecture_9 | ||
Wed 11/10 | Regularization | lecture_10 | ||
Fri 13/10 | Tree-based methods | lecture_11 | Xgboost | |
Wed 18/10 | Neural networks | lecture_12 | ||
Fri 20/10 | Neural networks for computer vision, Dropout, Batch Norm, ResNet | lecture_13 | Written_hw2 out | |
Wed 25/10 | Word embedding, RNN, LSTM | lecture_14 | ||
Fri 27/10 | Transformer | lecture_15 | ||
Wed 1/11 | NLP Pretraining, prompt | lecture_16 | ||
Fri 3/11 | Clustering | lecture_17 | Programming_HW2 out | |
Wed 8/11 | Limitations of deep learning: adversarial machine learning | lecture_18 | ||
Fri 10/11 | Semi-supervised learning, graph convolution network | lecture_19 | Graph laplacians | |
Wed 15/11 | Reinforcement Learning | lecture_20 | David Silver’s lecture | |
Fri 17/11 | AutoML(Neural architecture search) | lecture_21 | Homework3 out | |
Wed 22/11 | Review | lecture_22 | ||
Fri 24/11 | Final project presentation-part 1 | |||
Wed 29/11 | Final project presentation-part 2 |
References
There is no required textbook for this course. Some recommended readings are
- Deep Learning (by Ian Goodfellow, Yoshua Bengio, Aaron Courville)
- CS 229: Machine Learning, Stanford University