Machine Learning

COMP 5212, Fall 2024
The Hong Kong University of Science and Technology
Junxian He

Course description: this is an advanced machine learning course to provide a broad introduction to machine learning, covering foundational machine learning concepts, theory, algorithms, and applications. The topics include supervised learning (logistic regression, linear models, classification, SVM, kernal methods, naive Bayes), unsupervised learning (clustering, expectation maximization, graphical models, HMM), and reinforcement learning. It will include some optimization/learning topics (gradient descent, SGD, MLE and MAP estimation) as well. Certain aspects of deep learning are introduced(neural network basics, architectures, VAEs, GANs). The last lectures will give a general introduction to techniques in large language models.

Course Information

Course Logistics (Grading, Policy, etc.)

Time and Location

Lectures: Tue, Thu 1:30 PM - 2:50 PM at LG3008, Lift 10-12

Course Staff and Office Hours

Instructor: Junxian He. Office Hour: Tue 5 PM - 6 PM at Room 3512
TA1: Yuzhen Huang. Office Hour: Wed 9 AM - 10AM at the common place of SENG (near Lift 27/28, Room 2580)
TA2: Jinghan Zhang. Office Hour: Fri 6 PM - 7 PM at the common place of SENG (near Lift 27/28, Room 2580)

Contact and Communication

Canvas is the main platform for communication about technical questions of lecture contents and homeworks. Please do not directly email the teaching staff on technical questions which may easily get lost. The teaching staff will try to answer questions on Canvas in a prompt manner. Some rules:

  • All announcements and communications will happen over Canvas
  • Please utilize the Canvas Discussion page to ask questions. There are pre-existed discussion topics initiated by the staff such as lecture questions and each homework. Please post your questions in the respective thread, you are also encouraged to answer others’ questions if you know the answer.
  • While we allow students to optionally post anonymous questions and answers on Canvas, please do not post anything inappropriate as anonymous users – in such cases, we will turn off the anonymous option.
  • If you have non-technical, private matters, please email the teaching staff directly.
  • Please consult the Course Logistics page before asking logistical questions.
  • For longer discussions, please come to office hours.

Prerequisites and Materials

Students are required to be familiar with probability and linear algebra, and able to program well in Python. The CMU 10701 self-assessment exam is a good resource to roughly give you a notion of the background that is required for this course.

This course does not have a textbook, but here are some helpful materials:
Stanford CS229 Notes
Probability Review
The Matrix Cookbook
Linear Algebra Review

Lecture Schedule

The lecture schedule below is tentative and subject to change.

SlidesDateTopicReadingsAssignments
Lecture 003/09 TueIntroduction  
Lecture 1, annotated105/09 ThuMath basics  
Lecture 2, annotated210/09 TueLinear Regression  
Lecture 3, annotated312/09 ThuLogistic regression, Exponential Family HW1 Out
Lecture 4, annotated419/09 ThuGeneralized linear models, Kernel MethodsSection 3 of Notes 
Lecture 5, annotated523/09 MonSVMSection 5 of Notes 
Lecture 6, annotated624/09 TueSVMSection 6 of Notes 
Lecture 7, annotated726/09 ThuSVMSection 6 of Notes 
 01/10 TueThe National Day Holiday HW1 Due on 02/10
Lecture 8, annotated803/10 ThuGenerative ModelsSection 4 of Notes 
Lecture 9, annotated908/10 TueNaive Bayes, MLE, MAPSection 4.2 of Notes, Sec 2.1-2.3.6 of Mitchell Ch2, Section 4.2, 4.6 of probml 
Lecture 10, annotated1010/10 ThuGeneralization, bias-variance tradeoffSection 8.1-8.2 of Notes 
Lecture 11, annotated1115/10 TueClustering, EMSection 10, 11.1, 11.2 of NotesHW2 Out
Lecture 12, annotated1217/10 ThuExpectation MaximizationSection 11 of Notes 
Lecture 13, annotated1322/10 TuePCASection 12 of Notes 
Lecture 1424/10 ThuMid-term exam  
Lecture 15, annotated1529/10 TueProbabilistic Graphical ModelsKevin Murphy’s TutorialHW2 Due
Lecture 16, annotated1631/10 ThuHMMChapter 8, Speech and Language Processing 
Lecture 17, annotated1705/11 TueHMMChapter 8, Speech and Language Processing 
Lecture 18, annotated1807/11 ThuNeural Networks, backpropagationSection 7.4 of NotesHW3 out on 08/11
Lecture 19, annotated1912/11 TueNeural architectures Coding assignment out on 10/11
 14/11 ThuLecture Cancelled due to Typhoon  
Lecture 20, annotated2019/11 TueTransformer, Variational autoencoderThe Illustrated Transformer, The Annotated Transformer 
Lecture 21,annotated2121/11 ThuVariational autoencoderAuto-Encoding Variational BayesHW3 Due
Lecture 22,annotated2226/11 TueGANs, Reinforcement LearningGenerative Adversarial Networks, OpenAI Spinning Up in RLHW4 Out on 25/11
Lecture 2328/11 ThuLarge language models  
 2/12  HW4 Due
 8/12  Coding Assignment Due
 14/12Final Exam