course_machine_learning

Public repository for lecture notes / labs, etc

Course Logistics

Schedule

• Semester: Spring 2025
• Instructor: Tao LIN and Kaicheng YU
• Time and Location:
  • Theory: Tuesday 09:50 - 12:15, YunGu campus E10-212
  • Exercise: Thursday 08:00 - 09:35, YunGu campus E10-221
• Canvas

Grading

See course_info_sheet for more details.

Syllabus

Week	Session	Class Hour	Instructor	Theme / Topic	Teaching Activities (Lecture/Practical)
1st	Foundations of Data Science		Tao Lin	Course logistics; Introduction to ML (why ML, and why now); Linear algebra review; Probability review	Lecture
	Foundations of Data Science		TA	Lab 1 (graded): mathematical foundations	Practical
2nd	Linear Models for Regression		Tao Lin	Linear regression; Cost functions; Introduction to optimization	Lecture
	Linear Models for Regression		TA	Lab 2: Introduction to Python, NumPy, and PyTorch; Lab 1 due; Project 1 release	Practical
3rd	Linear Models for Regression		Tao Lin	Least squares; Probabilistic interpretation: Maximum Likelihood Estimation (MLE); Over- and under-fitting; Ridge regression; Lasso	Lecture
	Linear Models for Regression		TA	Lab 3	Practical
4th	Generalization, and Model Selection		Tao Lin	Generalization; Bias-Variance decomposition; Double descent phenomenon; Model selection, and validation	Lecture
	Generalization, and Model Selection		TA	Lab 4 (graded)	Practical
5th	Linear Models for Classification		Tao Lin	Classification; Logistic regression; Logistic regression and its optimization (MLE, Steepest descent, Newton's method, etc.)	Lecture
	Linear Models for Classification		TA	Lab 5; Lab 4 due	Practical
6th	Generalized Linear Models		Tao Lin	Exponential family; Generalized linear models	Lecture
	Generalized Linear Models		TA	QA for Project 1;	Practical
7th	Generative Learning Algorithms		Tao Lin	Discriminative vs. Generative learning algorithms; Gaussian Discriminant Analysis (GDA); GDA and Linear Discriminant Analysis (LDA); GDA and Naïve Bayes; GDA vs. Logistic regression	Lecture
	Generative Learning Algorithms		TA	QA for Project 1; Project 1 due; Project 2 release	Practical
8th	Nonparametric Methods		Kaicheng Yu	Parametric vs. nonparametric models; K-nearest neighbors; Decision trees; Bagging and random forest	Lecture
	Nonparametric Methods		TA	Lab 7	Practical
9th	Kernel Methods, SVM		Kaicheng Yu	Kernel methods; Support Vector Machine (SVM)	Lecture
	Kernel Methods, SVM		TA	Lab 6 (graded);	Practical
10th	Mixture Models, EM Algorithm		Tao Lin	Introduction to unsupervised Learning; Clustering; K-means; Gaussian Mixture Model (GMM); EM algorithm	Lecture
	Mixture Models, EM Algorithm		TA	Lab 6 due	Practical
11th	Neural Networks		Kaicheng Yu	Neural Networks - basics, representation power; Neural Networks – back-propagation	Lecture
	Neural Networks		TA	Lab 8	Practical
12th	Deep Neural Networks		Kaicheng Yu	Deep Neural Networks – advanced architectures (CNN, RNN, Transformer, etc.)	Lecture
	Deep Neural Networks		TA	Lab 9	Practical
13th	Deep Neural Networks		Tao Lin	Deep Neural Networks – optimization	Lecture
	Deep Neural Networks		TA	Lab 10 (graded); QA for Project 2	Practical
14th	Deep Neural Networks		Kaicheng Yu	Beyond supervised learning: - zero/few shot learning	Lecture
	Deep Neural Networks		TA	Lab 11 (graded); Lab 10 (due)	Practical
15th	Deep Neural Networks		Kaicheng Yu	Self-supervised Learning; LLMs	Lecture
			TA	Lab 12; Lab 11 (due); QA for Project 2	Practical
16th	Recitation		Tao Lin, Kaicheng Yu	Course recitation	Lecture
			TA	QA for Project 2; Project 2 due	Practical

Adjusted Syllabus for Exercise Sessions

Week	TAs (Head TA, *TAs)	Theme / Topic
13th	Shiwen LI, Peng SUN	Lab 1, Lab 3, Lab 4
14th	Peng SUN, Futing Wang	Lab 5, Lab 6, Lab 7
15th	Futing Wang, Enhui MA	Lab 8, Lab 9, Lab 10
16th	Enhui MA, Shiwen LI	Lab 11, Lab 12