Learning from Data (Fall 2021)
top
schedule
recitation
class project
Additional Links
fall 2020
home

Learning from Data (Fall 2021)

Welcome to the class website of Learning from Data!

Announcement

2021-12-30: The poster presentation information page has been added! Please make sure you checkout the poster requirement and grading policy.

2021-12-18: Please don't forget to complete the course evaluation on the Tsinghua Info site before Dec 24th and the TA evaluation before Dec 21st.

2021-12-12: PA5 is released! Due date: December 26th.

2021-11-25: Final project information is available! Please remember to fill in the team assignment form before Nov 30.

2021-11-19: PA4 is released! Due date: December 2nd.

2021-10-23: WA3 is released! Due date: November 26th. (An update is posted.)

2021-11-8 PA3 is updated with a correction. Please make sure to get the latest code!

2021-11-6 PA3 is released! Due date: November 20th.

2021-10-29: Midterm exam is next week! Off-campus students need to apply for taking the exam online with the TAs. The exam will be closed book and cover all content up to neural networks. You may bring an A4 double-sided cheat sheet to the exam. The midterm review session will be held on Saturday Oct 30th 10:00-12:00 am in Building C3-307.

2021-10-23: WA2 is released! Due date: November 3rd. No late day can be used for this homework.

2021-10-18: PA2 and WA1 updated for corrections.

2021-10-16: PA2 is released! Due date: Oct 28th

2021-9-17: Yang's office hour will be rescheduled to Wednesday, September 22nd 7-9pm.

2021-9-15: The review session will be held on Sunday, September 19th 2-5pm. Be sure to read the material and do the exercises before the review session！

Class info

  • Time: Friday 9:50-12:15

  • Location: C1-302

This introductory course gives an overview of many concepts, techniques, and algorithms in machine learning, beginning with topics such as logistic regression and SVM and ending up with more recent topics such as deep neural networks and reinforcement learning. The course will give the student the basic ideas and intuition behind modern machine learning methods as well as a bit more formal understanding of how, why, and when they work. The underlying theme in the course is statistical inference as it provides the foundation for most of the methods covered.

For more information about grading, homework and exam policies, see the class syllabus

Prerequisites： Undergraduate level calculus, probability and linear algebra. Basic Python programming.

Team

Instructor:

  • Yang Li <yangli@sz.tsinghua.edu.cn>

TAs

  • Weida Wang <wangwd19@mails.tsinghua.edu.cn>

  • Guoqing Zhang <wlsdzyzl@163.com>

  • Guodong Li <ligd19@mails.tsinghua.edu.cn>

Office hours:

Yang Friday 2:00-4:00pm Info Building 1108a
Guodong Thuesday 6:00-8:00pm Info Building,1111b
Weida Wednesday 6:00-8:00pm Info Building, 11th floor common area)
Guoqing Thursday 2:00-4:00pm (same as above)

You can also make appointments outside office hours.

Recitation Sessions

Recitations will be held every Friday 9:00-9:45pm in the lecture room.

Date Topic
Oct 16 Matrix derivatives, exponential models (recording) | (notes)

Oct 22 GDA & QDA , Multi-variate Bernoulli and Multi-nomial Naive Bayes (recording) | (notes) | (demo)

Oct 30 Midterm review (recording) | (review)

Nov 12 Midterm problem discussion (recording)

Nov 19 Review on eigen decomposition and SVD (notes) | (recording)

Nov 26 Tutorial on CNN (notes) | (recording)

Dec 3 Programming assignment 4 discussion
Dec 17 Written assignment 4 discussion (notes) | (recording)

Class Schedule

Date Topic Homework release
9/17 Introduction (slides) | (recording)

Review Session Exercises (WA0) See review notes

9/19 (2:00-5:00pm C1-302) Review Session I
(review handout) | (recording) | (programming review demo) | (math review notes)

Reference: The Matrix Cookbook

9/24 Supervised Learning I
(slides) | (slides with notes) | (recording)

Additional reading:

Programming Assignment 1 due Oct 7th.

10/8 Supervised Learning II: Multi-Class Classification
Generalized linear model
(slides with notes) | (recording)

Written Assignment 1 due Oct 22nd. updated on 10/18

WA1 solution

10/09 Supervised Learning III: Generative model (GDA)
(slides with notes) | (recording) | (code snippets)
10/15 Supervised Learning IV
Generative learning (Naive Bayes) (slides with notes)

Support vector machines
(slides with notes) | (recording)

Resources:

Programming Assignment 2 due October 28th. updated on 10/18

10/22 Supervised Learning V:
Soft margin SVM and Kernel SVM
Kernel regularized least square (slides with notes) | (recording)

Written Assignment 2 due November 3rd

WA2 solution

10/29 Neural Networks
(slides with notes - part 1) | (slides with notes - part 2) | (recording)

Resources:

Midterm practices (optional)

11/5 Midterm Exam Programming Assignment 3 due November 20th. updated on 11/08

11/12 Model selection
Regularization
Machine learning theory (part 1) (slides with notes) | (recording)

Written Assignment 3 due Nov. 26th

WA3 solution

11/19 Machine learning theory (part 2) (slides with notes)

Unsupervised Learning I: K-means clustering, spectral clustering
(slides with notes) | (recording)

Reference reading:

Programming Assignment 4 due Dec 2nd.

11/26 Unsupervised Learning II: spectral clustering, PCA
(slides with notes - part 1) | (slides with notes - part 2) | (recording)

Project information

Team sign up form due Nov 30th

Written Assignment 4 due Dec 12th

WA4 solution

12/3 Unsupervised Learning III: Kernel PCA, ICA and CCA
(slides with notes - part 1) | (slides with notes - part 2) | (recording) | PCA demo

Project proposal due on Dec 7th.
12/10 Reinforcement Learning
(slides with notes) | (recording)

Additional reading:

Programming Assignment 5 due Dec 26th.

12/17 Unsupervised Learning IV: EM & Factor Analysis
(slides with notes) | (recording)

Additional reading:

12/24 Semi-Supervised learning
(slides with notes) | (recording)

Additional reading:

12/31 Review and group discussion
1/4 Final poster session