Spring 2025 CS Course Listings

This file contains the Spring 2025 course listings for the Department of Computer Science and Engineering.

Archive of past courses

Course code: COMP 5111
Course title: Fundamentals of Software Analysis
Instructor: Prof. Shing-Chi Cheung
Room: 2534
Telephone: 2358-7016
Email:
WWW Page: https://course.cse.ust.hk/comp5111/

Area in which course can be counted: Software Engineering and Programming Languages (SEPL)

Course description:
This course aims to introduce the principles of various automated analysis and testing techniques as well as the ways they can be used to manage software code quality. Students will acquire fundamental knowledge of program abstraction, verification, testing, coverage, analysis, reliability, and fault detection. The course will also discuss how to carry out empirical experimentation for program analysis and testing. Wherever applicable, concepts will be complemented by industry adoption and tools developed in academia and industry. This enables students to understand the maturity and limitations of various techniques discussed in the course and put them into practice. The course requires prior programming knowledge in Java.

Course objective:
Students will attain the following on completion of the course:
* knowledge of the software quality assurance;
* understanding of principles of software testing and analysis;
* ability to deploy software quality measures to real life projects;
* hands-on experience in applying software testing and analysis tools.

Course outline/content (by major topics):
Software Testing, Program Analysis, Fault Diagnosis, Software Tool Automation, Large Language Models, Vulnerability Analysis, Empirical Experimentation

Textbooks:

Reference books/materials:
* Conferences: Proceedings of ICSE, FSE, ISSTA, ASE, PLDI, OOPSLA.
* Journals: ACM TOSEM & IEEE TSE.
* Software Engineering, Ivan Marsic, Rutgers University, 2012.
* Introduction to Software Testing, Paul Ammann and Jeff Offutt, Cambridge University Press, 2008.
* Software Testing and Analysis: Process, Principles and Techniques, Mauro Pezze and Michal Young, John Wiley and Sons, 2007.
* How Google Tests Software, James A. Whittaker, Jason Arbon, Jeff Carollo, Addison-Wesley, 2012.
* Head First Java, Kathy Sierra and Bert Bates, O'Reilly Media, Inc.

Grading scheme:
Programming Assignments: 20%
Reading Assignment: Report, Presentation & Participation: 10%
Examination: 70%

Available for final year UG students to enroll: Yes

Minimum CGA required for UG students: Permission of the instructor

Course code: COMP5112
Course title: Parallel Programming
Instructor: Dr. Qiong Luo
Room: 3511
Telephone: 2358-6995
Email:
WWW Page:

Area in which course can be counted: Software Engineering and Programming Languages (SEPL)

Course description:
Introduction to parallel computer architectures; principles of parallel algorithm design; shared-memory programming models; message passing programming models used for cluster computing; data-parallel programming models for GPUs; case studies of parallel algorithms, systems, and applications; hands-on experience with writing parallel programs for tasks of interest.

Exclusion(s): COMP 6111B, COMP 6511A, COMP 6611A

Background: COMP 3511 AND COMP 3711/COMP 3711H

Course objective:
Students will attain the following on completion of the course:
* knowledge of parallel computer architectures;
* understanding of principles of parallel algorithm design;
* knowledge of shared-memory and distributed-memory programming models;
* hands-on experience writing parallel programs of a task of interest.

Course outline/content (by major topics):
Introduction to parallel computer architectures;
principles of parallel algorithm design;
shared-memory programming models;
message passing programming models used for cluster computing;
data-parallel programming models for GPUs;
case studies of parallel algorithms, systems, and applications.

Reference books/materials:
* Introduction to Parallel Computing, Second Edition. Ananth Grama, George Karypis, Vipin Kumar, Anshul Gupta. Addison-Wesley, 2003, ISBN: 0201648652.
* Programming Massively Parallel Processors: A Hands-on Approach. Second Edition. David B. Kirk and Wen-mei W. Hwu. Elsevier 2012. ISBN: 978-0124159921.

Grading scheme:
project 50%, final exam 50%

Course code: COMP 5212
Course title: Machine Learning
Instructor: Dr. Junxian He
Room: CYT-3004
Telephone: 2358-8765
Email:
WWW Page: https://jxhe.github.io/teaching/comp5212s24

Area in which course can be counted: Artificial Intelligence (AI)

Course description:
This course covers core and recent machine learning algorithms. Topics include supervised learning algorithms (linear regression, logistic regression, generative models for classification, support vector machines), unsupervised learning (K-Means, mixture models, expectation maximization), deep learning, and reinforcement learning (classic RL, deep RL). The course assumes students have a solid grasp of probabilities, linear algebra, and python programming. The assignments and final projects will require proficient programming skills.

Background:
Computer science: object-oriented programming and data structures, design and analysis of algorithms; Mathematics: multivariable calculus, linear algebra and matrix analysis, probability and statistics.

Course objective:
Upon successful completion of the proposed course, students will be able to:
* Gain an overview of Machine Learning as a subject of study;
* Gain an understanding of the fundamental issues and principles in machine learning;
* Gain an understanding of core and recent machine learning algorithms;
* Gain an ability to apply core and recent machine learning algorithms to solve real-world problems.

Course outline/content (by major topics):
* Introduction to Machine Learning
* Basics of Probability/Information Theory
* Supervised Learning
    * Polynomial Regression and Basic ML Issues
    * Logistic Regression and Basic Optimization
    * Generative Models and Naive Bayes
    * The Bias-Variance Decomposition
    * Support Vector Machines
* Deep Learning
    * Deep Feedforward Networks
    * Convolutional Neural Networks
    * Recurrent Neural Networks
    * Transformer and BERT
* Unsupervised Learning
    * K-Means clustering
    * Expectation maximization
    * Variational autoencoders
    * Generative adversarial networks
* Reinforcement Learning
    * Introduction to RL
    * Value-Based Deep RL
    * Policy-Based Deep RL
* Large Language Models
    * Language Modeling
    * Prompting
    * Application

Reference books/materials:
o Andrew Ng. Lecture Notes on Machine Learning. Stanford. https://cs229.stanford.edu/syllabus.html
o I Goodfellow, Y Bengio, A Courville (2016). Deep Learning. MIT Press. https://www.deeplearningbook.org/

Workload and Grading:
o 4 Assignments: (40%)
o Final Group Project: (40%)
o Final Examination: (20%)

Available for final year UG students to enroll: PG students have higher priority to enroll in this course. UG students will be considered only if there are spare quota.

Minimum CGA required for UG students: N/A

Course code: COMP 5214
Course title: Advanced Deep Learning Architectures
Instructor: Dr. Qifeng Chen
Room: 3519
Telephone: 2358-8838
Email:
WWW Page:

Area in which course can be counted: Artificial Intelligence (AI)

Course description:
This course focuses on advanced deep learning architectures and their applications in various areas. Specifically, the topics include various deep neural network architectures with applications in computer vision, signal processing, graph analysis, and natural language processing. Different state-of-the-art neural network models will be introduced, including graph neural networks, normalizing flows, point cloud models, sparse convolutions, and neural architecture search. The students have the opportunities to implement deep learning models for some AI-related tasks such as visual perception, image processing and generation, graph processing, speech enhancement, sentiment classification, and novel view synthesis.

Course objective:
This course aims to achieve the following objectives:
* the students can have a broad knowledge of up-to-date advanced deep learning models in different areas;
* the students can utilize the deep learning architectures and techniques to solve the given programming assignments;
* the students can understand what problems can be addressed by deep learning models and solve a practical problem with deep learning through a course project;
* and the students can work in a team on a course project and present the work together.

Course outline/content (by major topics):
* Overview of deep learning: Basic architectures (CNN, RNN), Backpropagation, Loss functions
* Neural networks for image and video recognition tasks
* Neural networks for image and video processing tasks
* Deep 3D learning for point clouds, meshes, and volumetric data
* Deep 3D learning for stereo and multi-view data
* Graph neural networks for graph processing and analysis
* Sequential modeling and signal processing
* Deep generative models: Normalizing flow, GAN, Diffusion Models
* Efficient neural networks and Neural architecture search

Textbooks:

Reference books/materials:
o Zhang, A., Lipton, Z.C., Li, M. and Smola, A.J., 2019. Dive into deep learning. https://d2l.ai
o Goodfellow, I., Bengio, Y., Courville, A. and Bengio, Y., 2016. Deep learning (Vol. 1, No. 2). Cambridge: MIT press.

Workload and Grading:
o Presentation: 5%
o Homework: 30%
o Final project: 30%
o Midterm: 35%

Available for final year UG students to enroll: Yes

Minimum CGA required for UG students: Permission of the instructor

Course code: COMP 5221
Course title: Natural Language Processing
Instructor: Dekai Wu
Room: 3556
Telephone: 2358-6989
Email:
WWW Page:

Area in which course can be counted: Artificial Intelligence (AI)

Course description:
Language modeling from basics to LLMs. Techniques for parsing, interpretation, context modeling, generation. How neural and statistical approaches interact with linguistic constraints. Applications include machine translation, dialogue chatbots, cognitive modeling, and knowledge acquisition.

Background: COMP 3211

Exclusion(s): COMP 4221

Course objective:
(TBA)

Course outline/content (by major topics):
(TBA)

Textbooks:
(TBA)

Reference books/materials:
(TBA)

Grading scheme:
30% - 3 to 4 assignments
5% - participation (in class and on forum)
15% - midterm
20% - final
30% - project

Available for final year UG students to enroll: Yes

Minimum CGA required for UG students: Permission of the instructor

Course code: COMP 5311
Course title: Database Architecture and Implementation
Instructor: Prof. Xiaofang Zhou
Room: 3531
Telephone: 2358-7009
Email:
WWW Page:

Area in which course can be counted: Data, Knowledge and Information Management (DB)

Course description:
The course is divided in two parts: (i) background material taught by the instructor, (ii) specialized topics presented by students. In addition to their presentation, students have to submit a survey on the topic by the end of the semester.

Exclusion(s): COMP 3311

Course objective:
Introductory database class for graduate student that includes the relational model and SQL, disk and memory management, access methods, implementation of relational operators, query processing and optimization, transaction management and recovery. Moreover, students are expected to acquire presentation and writing skills.

Course outline/content (by major topics):
* E/R Model
* Relational Model and Algebra
* SQL
* Functional Dependencies and Relational Database Design
* File Systems
* Tree and Hash Indexes
* Query Processing and Implementation of Relational Operators
* Query Optimization
* Transactions

Textbooks:
Database System Concepts, A. Silberschatz, H. Korth, and S. Sudarshan.

Reference books/materials:
Database Management Systems, Raghu Ramakrishnan and Johannes Gehrke.

Grading scheme:
50% final exam, 20% presentation, 20% survey, 10% class participation

Available for final year UG students to enroll: No (UG students are encouraged to take COMP 3311)

Course code: COMP 5421
Course title: Computer Vision
Instructor: Prof. CK Tang
Room: 3538
Telephone: 2358-8775
Email:
WWW Page:

Area in which course can be counted: Vision and Graphics (VG)

Course description:
This course provides a comprehensive introduction to computer vision. Major topics include image processing, detection and recognition, geometry-based and physics-based vision and video analysis. Students will learn basic concepts of computer vision as well as hands on experience to solve real-life vision problems.

Background:
Instructor's approval is required for undergrad students to register the course for credit. We require that enrolled undergrad students should have GPA 3.7 (and grad students as well) or above because the course is programming intensive and require good mathematics skills.
Data structures and algorithms (COMP 3711)
Linear algebra (MATH 2111)
Matrix algebra and applications (MATH 2121)
A good working knowledge of Python programming

Course objective:
Same as listed in the course catalogue/academic calendar.

Course outline/content (by major topics):
1 Introduction
2 Image filtering
3 Image pyramids and (some) Fourier transform
4 Hough transform
5 Corner detection
6 Feature descriptors and matching
7 2D transformations
8 Image homographies
9 Camera models
10 Two-view geometry
11 Stereo
12 Structure from motion
13 Radiometry and reflectance
14 Radiometry continued
15 15 Photometric stereo and shape from shading
16 Image processing pipeline
17 Introduction to recognition
18 Bag of words
19 Neural networks
20 Convolutional neural network
21 Optical flow
22 Alignment and tracking
23 Correlation filters
24 Temporal models and SLAM
25 Graph-based methods

Textbooks:
Computer Vision: Algorithms and Applications, by Richard Szeliski.

Reference books/materials:
* Multiple View Geometry in Computer Vision, by Richard Hartley and Andrew Zisserman.
* Computer Vision: A Modern Approach, by David Forsyth and Jean Ponce.
* Dgital Image Processing, by Rafael Gonzalez and Richard Woods.

Grading scheme:
The breakdown is subject to change as a whole and adjustments on a per-student basis in exceptional cases. This is the general breakdown we'll be using:
Five programming assignments: 70%
Final Exam: 30%
Programming assignments are individual. Though you may discuss the problems with others, your answers must be your own.
There is no midterm. The final is comprehensive and is a closed-book-closed-notes exam.

Available for final year UG students to enroll: Yes

Minimum CGA required for UG students: A- and permission of the instructor

Course code: COMP 5422
Course title: 2D and 3D Deep Visual Scene Understanding
Instructor: Dr. Dan Xu
Room: 3509
Telephone: 2358-8837
Email:
WWW Page: https://www.danxurgb.net/

Area in which course can be counted: Vision and Graphics (VG) or Artificial Intelligence (AI)

Course description:
Visual scene understanding is a fundamental field for advanced application scenarios such as autonomous driving, AR/VR, and Embodied AI. This course mainly focuses on delivering deep learning-based visual scene understanding techniques in both 2D and 3D perspectives.

In the 2D perspective, we introduce topics including image and scene classification, semantic and instance segmentation, object detection, and multi-task scene perception and understanding. In the 3D perspective, we show how 3D scene understanding can be performed through learning from 2D/3D inputs, involving topics such as scene depth estimation, camera pose prediction, implicit and explicit 3D scene representations (e.g., NeRF and Gaussian Splatting), 3D scene reconstruction and generation, and visual SLAM. Several representative deep scene understanding architectures and frameworks in supervised or self-supervised settings, together with the 2D/3D tasks, are also presented in the course.

Background:
The instructor's approval is required for undergrad students to register for the course for credits. There are no strict prerequisites for this course. However, basic knowledge of computer vision and deep learning fundamentals is beneficial and necessary.

Course objective:
The objectives of the course are to help the students:
(i) Obtain basic knowledge of visual scene understanding techniques for various intelligent applications in autonomous driving, AR/VR, and embodied AI.
(ii) Learn fundamentals in deep learning-based architectures/frameworks for several important 2D and 3D visual scene understanding tasks.
(iii) Gain a sense of current research and development trends in academia and industry in the domain of visual scene understanding.
(iv) Learn skills of digging into and presenting for research papers published at top conferences/journals.

Course outline/content (by major topics):
Deep network architecture design and optimization, 2D and 3D scene perception and understanding tasks including 2D/3D object detection, semantic and instance segmentation, multi-task perception, scene depth and camera pose estimation, implicit and explicit 3D scene representations (NeRF and Gaussian Splatting), 3D scene reconstruction and generation, visual SLAM etc., and advanced visual scene understanding techniques for real-world large-scale application scenarios such as autonomous driving, AR/VR, and embodied AI.

Reference books/materials:
* Deep Learning, Ian Goodfellow and Yoshua Bengio and Aaron Courville, MIT Press, 2016.
* Multi-View Geometry in Computer Vision, Richard Hartley and Andrew Zisserman, Cambridge University Press, 2004.

Grading scheme:
Assignments 40%, Final project 30%, Final Exam 30%

Available for final year UG students to enroll: Yes

Minimum CGA required for UG students: 3.7 and permission of the instructor

Course code: COMP 5423
Course title: Deep Learning for Medical Image Analysis
Instructor: Dr. Hao CHEN
Room: 3524
Telephone: 2358-8346
Email:
WWW Page:

Area in which course can be counted: Artificial Intelligence (AI) or Vision and Graphics (VG)

Course description:
Nowadays medical image analysis is rapidly growing and plays an indispensable role in healthcare. Recent advances of deep learning techniques have made significant breakthroughs in medical image analysis applications. This course will cover fundamental knowledge of medical imaging and various medical image analysis tasks, including computer-aided detection, segmentation, diagnosis and prognosis. Deep learning methods for solving these tasks will be introduced and state-of-the-art methods will be discussed. The remaining significant challenges and limitations will also be presented, including limited amount of labeled data, deep learning with interpretation and generalization issues, etc. This course will equip students with practical knowledge of medical imaging and analysis with deep learning techniques.

Background:
Instructor's approval is required for undergraduate students to register the course for credits. Basic knowledge about image processing and machine learning are beneficial.

Course objective:
The objectives of the course are to help the students:
* Obtain the basic knowledge of medical imaging techniques and various medical image analysis tasks.
* Learn the fundamentals in deep learning methods for medical imaging and analysis.
* Master and apply the skills of deep learning technologies in medical image analysis tasks, including computer-aided detection, diagnosis and prognosis, etc.
* Gain the current research and development trends in both academia and industry in the domain of medical imaging and analysis.

Course outline/content (by major topics):
1. Introduction to Medical Image Analysis;
2. Fundamentals of Deep Learning;
3. Medical Image Classification;
4. Medical Image Segmentation;
5. Medical Image Registration;
6. Label-efficient Learning in MIA;
7. Anomaly Detection in MIA;
8. Attention Mechanism in MIA;
9. Interpretability in MIA;
10. Domain Generalization/Adaptation in MIA;
11. Federated Learning with Privacy-preserving;
12. Multimodal Learning in Healthcare;
13. Advances and Applications.

Reference books/materials:
Toennies, Klaus D. Guide to medical image analysis. Springer London, 2017.
Goodfellow, Ian, Yoshua Bengio, and Aaron Courville. Deep learning. MIT press, 2016.
Dhawan, Atam P. Medical image analysis. Vol. 31. John Wiley & Sons, 2011.
Zhou, S. Kevin, Hayit Greenspan, and Dinggang Shen, eds. Deep learning for medical image analysis. Academic Press, 2017.

Grading scheme:
Assignments 20%; Project Report and Presentation 50% and Final Exam 30%.

Available for final year UG students to enroll: Yes

Minimum CGA required for UG students: B

Course code: COMP 5631
Course title: Cryptography and Security
Instructor: Prof. Cunsheng Ding
Room: 2533
Telephone: 2358-7021
Email:
WWW Page: https://cse.hkust.edu.hk/faculty/cding/

Area in which course can be counted: Software Engineering and Programming Languages (SEPL)

Course description:
This course gives an in depth coverage of the theory and applications of cryptography, and system security. In the part about cryptography, basic tools for building security systems are introduced. The system security part includes electronic mail security, network security, Web security, Secure Shell, VPNs, and firewalls.

Course objective:
After completion of this course, students will display a breadth of knowledge of both the principles and practice of cryptography and systems security, master basic tools for building security systems, and get familiar with real-world security systems.

Course outline/content (by major topics):
History of cryptography, classical ciphers, design and analysis of block ciphers and stream ciphers, public-key cryptography, hash functions, keyed hash functions, digital signature schemes, user and data origin authentication, data integrity, nonrepudiation, Key management, public key infrastructure, cryptographic protocols, email security, Web security, network security, Secure Shell, VPNs, distributed systems security.

Textbooks:
No textbook, but lecture slides will be posted online.

Reference books/materials:
* W. Stallings, Cryptography Theory and Network Security, Fourth/Fifth Edition, Pearson Education.

Grading scheme:
Assignments and final examination.

Available for final year UG students to enroll: Yes

Minimum CGA required for UG students: A-

Course code: COMP 5712
Course title: Introduction to Combinatorial Optimization
Instructor: Sunil Arya
Room: 3514
Telephone: 2358-8769
Email:
WWW Page: https://cse.hkust.edu.hk/~arya/

Area in which course can be counted: Theoretical Computer Science (TH)

Course description:
* An introduction to the basic tools of Combinatorial Optimization.
* Includes: Linear Programming, Matching, Network Flow, Approximation Algorithms.

Background: COMP 3711 or equivalent + Linear Algebra

Course objective:
Upon completion of this course students will have been introduced to many of the most basic tools of combinatorial optimization and will be able to apply them towards designing efficient algorithms in their own research domains.

Course outline/content (by major topics):

Textbooks:

Reference books/materials:
Jiri Matousek and Bernd Gartner. Understanding and using linear programming, Springer, 2006.
Vijay Vazirani. Approximation algorithms, Springer, 2001.
David P. Williamson and David B. Shmoys. The design of approximation algorithms, Cambridge University Press, 2011
Jon Kleinberg and Eva Tardos. Algorithm design, Pearson/Addison-Wesley, 2006.

Grading scheme:
Homework, midterm and final examination.

Available for final year UG students to enroll: Yes

Minimum CGA required for UG students: Instructor Permission required.

Course code: COMP 5911 (co-list with COMP 4911/ENTR 4911)
Course title: Entrepreneurial Me
Instructor: Prof. Gary Chan
Room: 2539
Telephone: 2358-6990
Email:
WWW Page: https://course.cse.ust.hk/comp4911/

Area in which course can be counted: Networking and Computer Systems (NE)

Course description:
While entrepreneurship is a career choice, its mindset is for everyone. This is a course covering the mindset and elements of founding new and innovative business ventures in information technology sector. Topics include the entrepreneurial risk-taking value-creation mindset, market identification and go-to-market strategies, business models and development, business plan, fundraising and investment, role and protection of intellectual properties, technology-market gap and product-market fit, and growth and exit strategies. Case studies of successful and unsuccessful ventures will be discussed. In-class student participation and presentation are expected. Business and non-engineering students interested in starting IT-related companies are also welcome. Research postgraduate students are encouraged to develop proof-of-concept prototypes and business plans based on their research findings.

Background:
Any postgraduate students from all schools who are interested in starting up IT-related ventures. RPGs (MPhil and PhD candidates) of any majors with research topics are especially welcome.

Course objective:
After successful completion of the course students should know what is required to start a successful IT business. They will learn how to write a business plan to attract funding, the elements to form a winning team of people, understand and follow legal requirements, the steps required to move the company from the start-up phase to operating phase, and the factors to consider to exit and reap the rewards. They will also learn how to make succinct presentations of their ideas and communicate effectively in a team setting.

Course outline/content (by major topics):
* The entrepreneurial risk-taking value-creation mindset
* Business analysis and plan
* Business model development
* The role and protection of innovations
* VC, financing, fund raising and share structuring
* From 0 to exit: Building up your company
* Pitching and marketing your products
* Case studies and discussions

Reference books/materials:
N/A

Grading scheme:
3 Group projects (50%)
Lecture attendance and participation (~20%)
Individual reports (~15%)
In-class written exam (~15%)

Available for final year UG students to enroll: No. UG should enroll into COMP 4911/ENTR 4911

Minimum CGA required for UG students: N/A

Course code: COMP 6411C
Course title: Advanced Topics in Multimodal Machine Learning
Instructor: Dr. Long Chen
Room: CYT-3003
Telephone: 2358-8836
Email:
WWW Page: https://zjuchenlong.github.io/Teaching/

Area in which course can be counted: Vision and Graphics (VG)

Course description:
This course provides a comprehensive introduction to recent advances in multimodal machine learning, with a focus on vision-language research. Major topics include multimodal understanding (including translation, multimodal reasoning, multimodal alignment, multimodal information extraction), multimodal generation, multimodal pretraining and adaptation, and recent techniques and trends in multimodal research. The course structure will primarily consist of instructor presentation, student presentation, in-class discussion, and a course final project.

Background:
Machine learning basics, deep learning basics, computer vision basics

Course objective:
After completion of this course, students will understand mainstream multimodal topics and tasks, and develop their critical thinking and problem solving, such as identifying and explaining the state-of-the-art approaches for multimodal applications.

Course outline/content (by major topics):
Multimodal Understanding (Translation, Reasoning, Alignment, Information Extraction)
Multimodal Generation
Multimodal Pretraining and Adaptation
Recent Techniques and Trends in Multimodal Research

Reference books/materials:
Conferences: Proceedings of CVPR/ICCV/ECCV, ICLR/ICML/NeurIPS, ACL/EMNLP/ACM Multimedia
Book: Goodfellow, Ian, Yoshua Bengio, and Aaron Courville. Deep learning. MIT press, 2016.

Grading scheme:
In-class discussion: 20%
Project presentation: 30%
Final project report: 50%

Available for final year UG students to enroll: Yes

Minimum CGA required for UG students: 3.70

Course code: COMP 6611C
Course title: Advanced Topics in Embedded AI Systems
Instructor: Dr. Xiaomin Ouyang
Room: 3562
Telephone: 2358-8330
Email:
WWW Page: https://xmouyang.github.io/

Area in which course can be counted: Networking and Computer Systems (NE)

Course description:
This course will enable students to have an in-depth understanding of embedded AI algorithms and their implementation in real systems and applications. The major topics include 1) basics on machine learning; 2) data and system challenges in embedded AI 3) AI techniques and their implementation on cutting-edge platforms 4) real-world applications, such as smart health and smart buildings. The course structure will primarily consist of instructor presentations, student presentations, paper summaries, and a course project. Students will work on an individual or team project to build an end-to-end embedded AI system. Students will also read and discuss the latest publications in the areas of embedded AI, Internet of Things, mobile systems, and ubiquitous computing.

Background:
Basic understanding of machine learning is required. Familiarity or experience with the fundamentals of embedded or mobile systems are preferred.

Course objective:
After completion of this course, the students will have In-depth understanding of challenges in embedded AI systems, hands-on experience in implementing state-of-the-art embedded AI algorithms in real-world systems or applications, and critical thinking ability to tackle embedded AI problems.

Course outline/content (by major topics):
Machine Learning Basics
Challenges in Embedded AI Systems
Unsupervised Learning
Multimodal Sensing and Learning
Federated Learning
Efficient Deep Learning on the Edge
LLMs on the Edge
Physics-strengthened AI for Sensing Systems
Applications

Reference books/materials:
Conferences: Proceedings of MobiCom / MobiSys / SenSys / UbiComp / IPSN / IoTDI
Book: Siam, S.I., Ahn, H., Liu, L., Alam, S., Shen, H., Cao, Z., Shroff, N., Krishnamachari, B., Srivastava, M. and Zhang, M., 2024. Artificial Intelligence of Things: A Survey. ACM Transactions on Sensor Networks.

Grading scheme:
Participation and discussion: 20%
Project 60% (proposal@10+midterm presentation@10+final presentation@20+final report@20)
Paper presentation 10%
Paper Reviews 10%

Available for final year UG students to enroll: Yes

Minimum CGA required for UG students: Permission of the instructor

Please visit Class Schedule & Quota (Spring 2025) for the timetable and quota.

Archive of past courses

Last modified on 2025-01-23.