Spring 2006 CS Course Listings

This file contains the Spring 2006 course listings for the Department of Computer Science.

Archive of past courses

Course Code: COMP512
Course Title: Advanced Distributed Software Development

Instructor: S.C. Cheung
Room: 3554
Telephone: 23587016
Email:
WWW page: http://cse.hkust.edu.hk/~scc/

Area in which course can be counted: software technology

Course description (can be more detailed than the one in the calendar):
Introduction to selected advanced concepts of software development in distributed environments. Topics include theoretical models and analysis, object-oriented methodologies for distributed applications, concepts and technologies of software applications. The course will include programming and/or reading assignments.

Course objective:
To understand the underlying principles of distributed software systems and to adopt a rigorous approach in the development of such systems.

Course outline/content (by major topics):

  • Object-oriented methodologies
  • Architecture - Meeting today's challenges
  • UML for enterprise software development
  • Object design patterns and software evolution
  • Fundamental concepts and models

Text book: n/a

Reference books/materials: Please refer to the course web site.

Grading Scheme:

  • Participation (10%)
  • Assignments (40%)
  • Project (50%)

Background needed: software engineering, Java, database

Available for final year UG students to enroll: No.

Course Code: COMP526 ( Cancelled )
Course Title: Natural Language Processing

Instructor: Dekai Wu
Room: 3539
Telephone: 2358-6989
Email:
WWW page: http://cse.hkust.edu.hk/~dekai/

Area in which course can be counted: AI

Course description:
The advent of Google and the intense competition it has inspired among Microsoft, Yahoo, IBM and others, has drawn attention to the clear emergence of NLP as one of this decade's highest-impact areas of computer technology in the real world as well as in research. NLP has become the key technology that enables much of Intelligent Search, Knowledge Management, Data Mining, Document Management, Speech Recognition, the Semantic Web, and more. According to industry expert studies by the Gartner Group, NLP technologies represent HALF of the 12 most important growth areas of computer technology during 2001-2010. It is difficult to exaggerate the importance of this core information infrastructure technology.

The excitement behind this area will be investigated in this course. The field has grown so quickly that it can be difficult to gain an up-to-date perspective. We will explore application of both statistical modeling techniques (for which the Human Language Technology Center at HKUST is particularly recognized) and symbolic knowledge-based modeling techniques. We will share insights from both the theoretical perspectives as well as the applied perspectives.

NLP is a rich, challenging, and rewarding field that sits at the intersection of science, technology, and psychology. It is for those who value a broad, wide-ranging perspective but are not afraid of drilling deep at the same time. I look forward to meeting those of you who are interested in such directions.

Course objective:

  • To establish a broad perspective over the foundations of NLP.
  • To understand the methods, issues, and techniques via case studies.
  • To learn hands-on how to turn theory to application.

Course outline/content (by major topics):
NLP is an extremely broad area. Specific key topics and cases across lexical, syntactic, semantic, and contextual processing will be determined according to the class composition.

Text book:
Foundations of Statistical Natural Language Processing, by Christopher D. Manning & Hinrich Schutze.

Reference books/materials:

  • Handbook of Natural Language Processing, edited by Robert Dale, Hermann Moisl, & Harold Somers.
  • Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics and Speech Recognition, by Daniel Jurafsky & James H. Martin.
  • Readings and cases

Grading Scheme:

  • 35% Class participation
  • 65% Project

Background needed: COMP221; Background in any AI or statistical areas will help you get more out of the class.

Available for final year UG students to enroll: Instructor's consent.

Minimum CGA required for UG students: Instructor's consent.

Course Code: COMP 537
Course Title: Knowledge Discovery in Databases

Instructor: Qiang Yang
Room: 3563
Telephone: 8768
Email:
WWW page: http://cse.hkust.edu.hk/~qyang/

Area in which course can be counted: Databases

Course description:
Data mining has emerged as a major frontier field of study in recent years. Aimed at extracting useful and interesting patterns and knowledge from large data repositories such as databases and the Web, data mining has successfully integrated techniques from the fields of database, statistics and AI. This course will provide a broad overview of the field, preparing the students with the ability to conduct research and development in the field.

Course objective:
To learn the most recent developments in data mining and machine learning research and practice. To help the students get ready for research.

Course outline/content (by major topics):

  • The Data Mining Process
  • Preprocessing and Model Evaluation
  • Supervised and Semi-supervised Learning
  • Unsupervised Learning
  • Utility and Cost-sensitive Learning
  • Applications: Web and Text Data Mining
  • Applications: Ranking and CRM application, Security applications, Bioinformatics
  • Applications: Object tracking in Wireless Networks

Text book:

  • Data Mining -- Practical Machine Learning Tools and Techniques with Java Implementations by Ian Witten and Eibe Frank, Morgan Kaufmann Publishers.
  • Data Mining -- Concepts and Techniques by Jiawei Han and Micheline Kamber. Morgan Kaufmann Publishers.

Reference books/materials:

Grading Scheme:

  • Midterm Exam 30%
  • Homework Assignments and Projects 30%
  • Final Exam 40%

Background needed: COMP231

Available for final year UG students to enroll: Yes, but by approval.

Minimum CGA required for UG students: permission of the instructor

Course Code: COMP 573
Course Title: Computational Geometry

Instructor: Sunil Arya
Room: 3509
Telephone: 2358-8769
Email:

Area in which course can be counted: Theory

Course description (can be more detailed than the one in the calendar):
This is an introductory course in Computational Geometry. It deals with the design and analysis of algorithms for geometric problems.
Examples of objects to be studied include Convex hulls, Voronoi
diagrams, and Triangulations.

Course outline/content (by major topics):

  • Convex Hulls
  • Line Segment Intersection
  • Polygon Triangulation
  • Linear Programming
  • Orthogonal Range Searching
  • Point Location
  • Voronoi Diagrams
  • Arrangements and Duality
  • Delaunay Triangulations

Text book:
M. de Berg, M. van Kreveld, M. Overmars and O. Schwarzkopf, Computational geometry---algorithms and applications, Springer-Verlag, 2000.

Grading Scheme:

  • 3-5 written assignments 30%
  • Midterm exam 30%
  • Final exam 40%

Background needed: COMP271

Available for final year UG students to enroll: Yes.

Mimimum CGA required for UG students: Permission of instructor required.

Course Code: COMP621N
Course Title: Advanced Topics in AI: Decoding in Statistical Machine Translation

Instructor: Dekai Wu
Room: 3539
Telephone: 2358-6989
Email:
WWW page: http://cse.hkust.edu.hk/~dekai/

Area in which course can be counted: AI

Course description (can be more detailed than the one in the calendar):
Statistical machine translation research typically focuses mainly on the learning phases, in areas such as alignment and translation model training. But the accuracy of statistical machine translation models are also strongly limited by their decoders, which are responsible for heuristically exploring the search space for an optimum translation.
In this course, we will explore characteristics of the search space and their impact on various aspects of decoding and decoder architectures.

Course objective:

  • To explore various approaches to addressing decoding weaknesses in SMT (statistical machine translation) models.
  • To understand the methods, issues, and techniques via case studies.
  • To learn hands-on how to turn theory to application.

Course outline/content (by major topics):

  1. Empirical error analysis on current state-of-the-art SMT decoders
  2. Approaches to addressing decoder deficiencies
  3. Rescoring and reranking approaches
  4. Ensemble techniques, ROVER (Recognizer Output Voting Error Reduction)
  5. Comparative analysis
  6. Design, implementation, and evaluation of new models

Text book: Readings and cases.

Reference books/materials:

  • The Theory of Parsing, Translation, and Compiling (Volumes 1 &2), by Alfred V. Aho and Jeffrey D. Ullman. (1972)
  • Foundations of Statistical Natural Language Processing, by Christopher D. Manning & Hinrich Schutze. (June 1999)
  • Handbook of Natural Language Processing, edited by Robert Dale, Hermann Moisl, & Harold Somers. (July 2000)
  • Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics and Speech Recognition, by Daniel Jurafsky & James H. Martin. (Jan 2000)

Grading Scheme:

  • 35% Class participation
  • 65% Project

Pre-requisites/Background needed:

  • Instructor's consent required.
  • Background equivalent to COMP526 (Natural Language Processing).
  • Background equivalent to COMP621H (Machine Translation).
  • Background in any AI or statistical areas will help you get more out of the class.

Available for final year UG students to enroll: No.

Minimum CGA required for UG students: N/A

Course Code: COMP 630J
Course Title: Topics in Database Systems: Similarity Search Over Databases

Instructor: Lei Chen
Room: 3546
Telephone: 2358 6980
Email:
WWW page: http://cse.hkust.edu.hk/~leichen/

Area in which course can be counted: database

Course description (can be more detailed than the one in the calendar):
This course covers topics related to similarity search over database, such as, distance function design, indexing methods, and dimensionality reduction techniques. After introducing the basic techniques related to similarity search, projects will be assigned to students. The projects are related to similarity search over multimedia, stream and network data.

Course objective:
The objective of this course is to give students an overview of similarity search over database and let students get familiar with the basic techniques related to similarity search. Through the projects, students will learn how to start a research in this area.

Course outline/content (by major topics):

  1. Overview of similarity search
  2. Distance functions
  3. Data representations
  4. Dimensionality reductions
  5. Indexing methods
    5.1 Indexing over vector space
    5.2 Indexing over metric space
    5.3 Indexing over non-metric space
  6. Similarity search over network data

Text book: No.

Reference books/materials:

Grading Scheme:

  • Class Presentation 20%
  • Discussion Participation 20%
  • Class Attendance 10%
  • Project Report 50%

Pre-requisites/Background needed: None

Available for final year UG students to enroll: No.

Minimum CGA required for UG students: N/A

Course code : COMP 641L
Title : Topics in Graphics: Fundamentals in 3D Computer Vision

Instructor: Long QUAN
Room: 3506
Email: quan
Web page :

Area in which course can be counted: VG

Course description:
This course will cover some fundamental topics of modern three-dimensional computer vision. It includes some background introduction of geometry and optimization, the geometry of multiple views, structure from motion,3D reconstruction and modeling. It will also discuss feature detection and image segmentation to build up high-level representation of 3D objects and scenes.

Course objective:
Understanding the fundamentals of modern computer vision approaches, particularly in the area of 3D computer vision.

Course outline/content (by major topics):
Introduction to geometry, study of the geometry of multiple views, feature detection from images, statistical and optimization tools, image-based modeling and rendering, future directions of modeling

Text book: none

Reference books/materials: later on

Grading Scheme: projects and presentations

Background needed:
Some basic math background in linear algebra and numerical analysis is desirable.

Exclusion (if applicable): no

Available for final year UG students to enroll: permission by instructor

Minimum CGA required for UG students:

Course code : COMP 660J
Course Title: Topics in Computer & Communication Networks: Hot Topics in Computer and Internet Communications

Instructor: Brahim Bensaou
Room: 3525
Telephone: 23587014
Email: ;
WWW page: http://cse.hkust.edu.hk/~csbb/

Area in which course can be counted: Computer Engineering

Course description:
Most of today’s network services require a reliable transport of elastic data traffic. Such traffic is also forecast to grow steadily over the coming years. At the heart of such transport lies the well known Transmission Control Protocol (TCP). One of the major functions of TCP is to deploy and end-to-end congestion control mechanism with the objective of preventing congestion or mitigating its effects. In the ever evolving Internet, new network technologies – viz. the Web, the increase of link bandwidth, the access of data over long distances, and over error prone wireless links, and so on – are today challenging the ability of TCP to maintaining a good network performance. In this course, through lectures, paper readings, and presentations, we will get familiar with such challenges and problems that face TCP, and study recent advances and solutions to some of these problems when they exist.

Course objective:
To present state-of-the-art research problems on congestion control in modern computer networks and study some of the solutions proposed in recent papers. Through paper readings, review, and presentation, students will gain a broad understanding of the current technologies and research efforts in Internet traffic control. The students will also be able to gain some experience in preparing research proposals and undertaking some research in the chosen area.

Course outline/content (by major topics):

  • End-to-end Congestion control
    o TCP Flavors (Tahoe, Reno, Vegas)
    o TCP models
    o TCP performance evaluation
  • TCP and Active Queue Management (AQM)
    o RED, Choke, PI, REM, …
    o Interaction between TCP and AQM
  • Router assisted congestion control,
    o XCP, RCP and their performance
  • TCP in high bandwidth-delay product networks
    o TCP FAST, TCP Westwood, …
  • Available Bandwidth Inference in the Internet
    o Packet pairs principle
    o IGI, PathChirp, PathLoad, …
  • Wireless Networks and Congestion control
    o TCP for wireless networks: I-TCP, M-TCP, TCP Veno
    o Ad-hoc Networks and congestion control

Reference books/materials:
Material: recent research papers to be distributed on the course webpage

Grading Scheme:

  • Participation (10%)
  • Paper Review (10%)
  • Class Presentation of papers (25%)
  • Project Report (50%)
  • 15% Bibliographic search and state of the art review
  • 10% Project proposal,
  • 20% Project final report and originality
  • 20% Final project presentation

Background needed:
Students should have taken a first course in computer networks equivalent to Comp 361 and must have some basic mathematical knowledge and analytical skills

Available for final year UG students to enroll: No

Course code : COMP 670O
Course Title: Topics in Theory: Game Theoretic Applications in CS

Instructor: Mordecai Golin
Room:
Telephone: 23586993
Email:
WWW page: http://cse.hkust.edu.hk/~golin/

Area in which course can be counted: No area

Course description:
A quick and dirty introduction to the basic concepts of modern game theory followed by a survey of recent work in computer science that uses game theory.

Course objective:
To introduce students to a fast-growing research field in modern computer science.

Course outline/content (by major topics):
The course will start with the instructor giving two or three introductory lessons on the basics of game theory. The remainder of the course will be students presenting papers. All students will be expected to

  • Read ALL of the papers assigned
  • Present two papers

Tentative topics to be covered are:

  • Introductory game theory
  • Game theory in ad-hoc networks
  • Game Theory and the internet
  • Combinatorial auctions
  • Evolutionary game theory and repeated games

Text Book:
No textbook.
We will be reading recent papers.

Reference books/materials:

Grading Scheme: TBA

Pre-requisites/Background needed:
COMP271 or equivalent and mathematical sophistication.
Enrollment is only with permission of the instructor.

Available for final year UG students to enroll: No

Course code : COMP 680H
Course Title: Topics in Computer Engineering: Advanced Topics in Next-Generation Wireless Networks

Instructor: Qian Zhang
Room: 3533
Telephone: 23588766
Email:
WWW page: http://cse.hkust.edu.hk/~qianzh/

Area in which course can be counted: Computer Engineering

Course description:
Wireless networks have gone through an unprecedented growth in past few years and will continue to play an indispensable role in future communications. A number of new wireless networking paradigms are emerging that might have significant impact on the next-generation wireless networks. In this course, through lectures, paper readings, and presentations, you will learn the recent advances in wireless networking research; we will study, discuss and criticize a number of research covering areas including: WiFi, WiMAX, and UWB, ad hoc networks, mesh networks, multi-radio networks, and open spectrum. In this course, you will be exposed to basic networking principles and how they are applied in real systems, and we also explore research issues and opportunities in medium access, topology control, routing, transport protocols, radio resource management and spectrum management.

Course objective:
To present state-of-the-art in wireless networking research. Through paper readings, review, and presentation, students will gain a broad understanding of the current technology and research efforts in wireless networking. In addition, students are expected to accomplish a well-defined research project in a group of 1-3 to obtain experience in designing and evaluating protocols and/or techniques for wireless networks.

Course outline/content (by major topics):

  • New wireless networks: WiFi, WiMAX, and UWB networks
  • Ad hoc networks
    o MAC, network, and transport protocols
    o Cross-layer design
  • Mesh networks
  • Multi-radio network systems
  • Open spectrum

Reference books/materials:
Material: recent research papers to be distributed in class or course webpage

Grading Scheme:

  • Participation (10%)
  • Paper Review (10%)
  • Class Presentation of papers (25%)
  • Project Presentation (15%)
  • Project Report (40%)

Background needed:
Students should have some knowledge in computer networks and operating systems

Available for final year UG students to enroll: No

Course Code: COMP 680 I
Course Title: Topics in Computer Engineering: Peer-to-Peer Computing

Instructor: Yunhao Liu
Room: 3548
Telephone: 2358-7019
Email:
WWW page: http://cse.hkust.edu.hk/~liu/

Course description:
P2P communication model established in overlay networks is an emerging technology aiming to effectively utilize and manage increasingly large and globally distributed information and computing resources in Internet, complementing existing client-server services. In order to truly adopt the P2P model for widely deploying large-scale Internet applications, and timely merge this model as an indispensable component in main stream distributed computing systems, we must address several major technical challenges including the efficiency and scalability of overlay networks, cost-effective and fast P2P information search, and privacy and security protection of peers.

Course objective:

Course outline/content (by major topics):

  1. Why P2P?
  2. A brief history and Evolution of P2P;
  3. P2P architectures;
  4. Usage of P2P technology;
  5. Trustworthy P2P;
  6. P2P observations.

Grading Scheme:

  • Class Presentation 20%
  • Discussion Participation 20%
  • Class Attendance 10% at most three misses
  • Project Report 40% Date TBD

Pre-requisite: COMP 561

Registration requirement: Instructor’s approval is needed

Course Code: COMP 685B
Course Title: Topics in Applications of Computer Science: Computer Music

Instructor: Andrew Horner
Room: 3537
Telephone: 2358-6998
Email:
WWW page: http://cse.hkust.edu.hk/faculty/horner/comp685b/

Area in which course can be counted: Computer Applications

Course description:
The techniques for synthesizing music in software synthesis, soundcards, hardware synthesizers, and mobile phones. Topics include: music representation, music theory, musical acoustics, spectral analysis, sound synthesis, sound modification techniques and effects, MIDI, music information retrieval, musical instrument recognition, music on mobile phones. No previous musical background required.

Course objective:
To learn the techniques for synthesizing music in software synthesis, soundcards, hardware synthesizers, and mobile phones. Course outline/content (by major topics): Music representation, music theory, musical acoustics, spectral analysis, sound synthesis, sound modification techniques, effects, MIDI, musical instrument recognition, pitch detection, signal separation, music on mobile phones.

Text book: none

Reference books/materials:

  • Computer Music Tutorial by Curtis Roads, MIT Press, 1996
  • Computer Music by Charles Dodge and Thomas Jerse, Schirmer Books, 1997 (2nd Ed)
  • Cooking with Csound by Horner & Ayers, A-R Editions, 2002

Grading Scheme:
2 Hands-on-experience assignments (20%), Midterm (40%), Project and Presentation (40%)

Pre-requisites/Background needed: None

Available for final year UG students to enroll: Yes

Minimum CGA required for UG students: permission of the instructor

Course Code: COMP696S
Course Title: Independent Studies: Information Visualization

Instructor: Huamin Qu
Room: 3508
Telephone: 2358-6985
Email:
WWW page: http://www.huamin.org/

Area in which course can be counted:

Course description:
Information visualization is emerging as a hot research field which fuses computer graphics, cognitive psychology, data mining, graphic design, and human-computer interaction. This course aims to provide an in-depth view of some advanced topics in information visualization. Topics include human perception, graph drawing algorithms, multivariate data encodings, network visualization, and user studies. The coursework will include readings, seminar presentations, and a real information visualization project.

Course objective:
To provide an introduction to information visualization; to gain an in-depth view of some advanced topics in information visualization.

Course outline/content (by major topics):

  1. Human Perception
  2. Treemaps
  3. Graph Drawing Algorithms
  4. Network Visualization
  5. Multivariate Data Visualization
  6. User Studies

Text book:

  • Collin Ware, "Information Visualization: Perception for Design", 2nd Edition, Morgan Kaufman Publishers/ 2004
  • Chaomei Chen, "Information Visualization: Beyond the Horizon", 2nd Edition, Springer/2004

Reference books/materials: Research papers

Grading Scheme:

  • 30% Discussion
  • 20% Presentations
  • 50% Project

Pre-requisites/Background needed: COMP341 (Computer Graphics)

Available for final year UG students to enroll: No

Please visit https://www.ab.ust.hk/wcr/cr_class_staf_main.htm for the timetable and quota.

Archive of past courses

This web page was created by Gang Zeng on 12 Dec. 2005.

Last modified on 06 Feb. 2006.