Scribd
Upload
40 views2 pages

Statistical Learning 2024 Syllabus

Uploaded by

luticia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
40 views2 pages

Statistical Learning 2024 Syllabus

Uploaded by

luticia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

COMP3354 Statistical Learning

Administrative
Instructor in Charge: Yingyu Liang
Instructor email: yingyul@hku.hk

TA: Yujin Han


Instructor email: yujinhan@connect.hku.hk

Course Description
The challenges in learning from big and complicated data have led to significant advancements
in the statistical sciences. This course introduces supervised, unsupervised learning, and reinforce-
ment learning, with emphases on the theoretical underpinnings and on applications in the statistical
programming environment R. Topics include Naïve Bayes, linear methods for regression and clas-
sification, basic expansions and regularization, model assessment, kernel methods, cluster analysis,
principal component analysis, and reinforcement learning. We will also provide an introduction to
neural networks and some applications.
This course is for students who would like to dive deep into the mathematical foundations of
machine learning and/or plan to research statistical learning. This course aims to provide them with
the mathematical foundations of machine learning and prepare them for more advanced study and
research in machine learning and AI. The course will focus on the design principles of machine
learning methods (where they come from) and the analysis principles (how they end up).

Prerequisites
MATH1853 or MATH2101 or STAT1602 or STAT1603 (i.e., undergraduate linear algebra, statistics,
and probability). Strong backgrounds in linear algebra, statistics, and probability are highly recom-
mended, since the course focuses on the mathematical foundations of machine learning. The course
is open to junior/senior undergraduates with the required math/statistics background.

References and Textbooks


There are no required textbooks. But the following books are good references. (Note: only some
parts of the references are used; the scope of the course is defined by the lectures, not by these
references.)
• Statistical Inference (2nd Edition). George Casella, and Roger L. Berger.
• Machine Learning: A Probabilistic Perspective. Kevin Murphy. Link
• Understanding machine learning: From theory to algorithms. Shai Shalev-Shwartz, and Shai
Ben-David. Cambridge University Press, 2014. Link
• High-Dimensional Probability: An Introduction with Applications in Data Science. Roman
Vershynin. Cambridge University Press, 2018. Link

Website and Computing Tools


We normally will use the HKU Moodle website to post lecture materials, homeworks, data, code
examples, etc.

1
Grading Policy
The course will be graded based on Continuous Assessment (50%, homework + potentially midterm
exam) and Written Examination (50%, final exam).
Homeworks: There will be about 5 bi-weekly homeworks for the first 10 weeks. Each homework
will contain mostly written questions, and potentially have some programming questions.

Tentative Syllabus
This syllabus is suggestive and may be updated slightly according to the teaching plan and student
backgrounds. Review classes/midterm may be inserted between the regular lectures.

• Introduction: Overview of machine learning


• Supervised learning: Review of probability/statistics, Naïve Bayes

• Supervised learning: Linear methods for regression and classification


• Supervised learning: Basic expansions and regularization
• Supervised learning: Model Assessment: Bias-variance trade-off, Generalization bounds

• Supervised Learning: Kernel methods


• Unsupervised learning: Cluster analysis
• Unsupervised learning: Principal component analysis, (Autoencoders)
• Reinforcement learning: Reinforcement learning and MDP

• Advanced topics: Neural Networks, Applications


• Course overview and wrap up
• Final exam

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy