Computer Science Program
Overview of the Program
Computer Science at KAUST offers Master and PhD degrees. Both graduate degrees require course work. Each student has a faculty member as advisor/supervisor, who can provide advice on course selection and directions for research. Computer Science offers five tracks, each of which leads to a frontier of computing:
- Theoretical Computer Science (Theory)
- Computer Systems (Systems)
- Artificial Intelligence (AI)
- High Performance Computing
- Visual Computing
For the Master degree, a student can chose one of the following directions:
- Master degree without thesis option: minimum of 30 credit hours of courses, or
- Master degree with thesis option: minimum of 30 credit hours of courses and a thesis.
Thesis credit-hours range from 6 to 12. A student must choose between the options before starting the second term of study.
For the PhD degree, a student must:
- Complete a minimum of 6 credit hours 300-level courses (after a Master’s degree), or additional 30 credit hours of course work for students bypassing the Master program.
- Complete the PhD qualification and candidacy phases.
- Present two seminars or lectures.
- Submit an annual progress report.
- Submit a thesis, which contains the candidate’s scholarly work. Thesis work must be publishable in well recognized journals and conferences.
Complete a thesis defense oral examination. These basic requirements are explained in 4.
1.1 Advisors/Supervisors
Each Master student is assigned to a faculty member who is responsible for advising the student on the selection of courses and the options available. The advisor also signs the student’s forms. After the first term of registration, the student may select the same or a different faculty member (with the agreement of the faculty member) as research supervisor for the thesis option only. A PhD student must have a research supervisor assigned prior the admission to the program. The research supervisor also serves as the advisor. Every student has the right to select co-supervisors from within the division. A maximum of 2 co-supervisors is permitted to supervise a graduate student. Each supervisor or co-supervisors must have an academic faculty-level appointment.
1.2 Program length
For the Master degree without thesis, a minimum of one year of full-time registration (4 courses, 12 credit hours) is normally required. Master degree with thesis option requires up-to four semesters. The minimum period of registration for the PhD degree is five semesters after a Master’s degree (or seven semesters after a Bachelor’s). The actual length of the PhD program depends on the student’s preparation and choice of research topic.
1.3 Progress Reports
Progress reports are required of Master students under the thesis option and of PhD students. They are intended to assist the student to focus on making timely progress through the program requirements. Students are required to submit annual progress report to the supervisor. The student progress can be discussed among the student advisory committee.
1.4 Courses
Graduate courses are classified into two levels: 200-level courses which are basic graduate courses, 300-level courses which are research-oriented courses. The program offers the following course classified as Core, Track, and Doctoral and Master level research courses.
Core courses:
- (CS221) Artificial Intelligence
- (CS 240) Operating Systems and Systems Programming
- (CS 242) Programming Languages
- (CS260) Design and Analysis of Algorithms
- (CS282)Computer Architecture and Organization Immersion
courses in each track:
Theoretical Computer Science
- (CS 212) Linear and Nonlinear Optimization
- (CS 229) Machine Learning
- (CS 248) Computer Graphics
- (CS260) Design and Analysis of Algorithms
- (CS261) Algorithmic Paradigms
- (CS 361) Combinatorial Machine Learning
- (CS 377) Information Networks
Computer Systems
- (CS 244) Computer Networks
- (CS 245) Databases
- (CS 248) Computer Graphics
- (CS282) Computer Architecture and Organization
- (CS341)Advanced Topics in Data Management
- (CS344)Advanced Topics in Computer Networks
- (CS346)Advanced Topics in Operating Systems
- (CS 380) GPU and GPGPU Programming
Visual Computing
- (CS 248) Computer Graphics
- (CS 247) Scientific Visualization
- (CS 272) Geometric Modeling
- (CS271) Applied Geometry
- (CS 380) GPU and GPGPU Programming
Artificial Intelligence
- (CS 212) Linear and Nonlinear Optimization
- (CS 229) Machine Learning
- (CS 245) Databases
- (CS 340) Computational Methods in Data Mining
- (CS341)Advance Topics in Data Management
- (CS 361) Combinatorial Machine Learning
High Performance Computing
- (CS282) Computer Architecture and Organization
- (CS 291) Scientific Software Engineering
- (CS 292) Parallel Programming Paradigms
- (CS 311) High Performance Computing I
- (CS 312) High Performance Computing II
- (CS 380) GPU and GPGPU Programming
Master-level Research
- (CS 298) Master Graduate Seminar
- (CS 299) Master Directed Research
- (CS 297) Master Dissertation Research
Doctoral-level Research
- (CS 398) Doctoral Graduate Seminar
- (CS 399) Doctoral Directed Research
- (CS 397) Doctoral Dissertation Research
- (CS341)Advanced Topics in Data Management
- (CS344)Advanced Topics in Computer Networks
- (CS346)Advanced Topics in Operating Systems
- (CS 380) GPU and GPGPU Programming
- (CS 361) Combinatorial Machine Learning
2.0 Master Degree without Thesis Requirements
For the Master degree without thesis option, course requirements include four Core courses, three Immersion courses, one course from the student field of research (e.g., one immersion course or a directed research), one cognate course, and one elective course. Detailed description of courses is available in Section 5.
3.0 Master Degree with Thesis Requirements
For the Master degree thesis option, course requirements include four Core courses, three Immersion courses, one course from the student field of research (e.g., one immersion course or a directed research), one cognate course, and one elective course. Detailed description of courses is available in Section 5 . An additional six to twelve thesis credit hours (CS 297 Dissertation Research) are required. By the end of the second semester, a student must have arranged a research supervisor and agreed on the general area of the proposed research. Before the end of the third semester, the student must have defined a thesis topic in consultation with the supervisor. Two readers for the thesis, in addition to the supervisor, must also be chosen. The mutual agreement of the student, supervisor, and readers must be reported in writing and kept at the division. The agreement must explicitly state the topic of the thesis, and the expected completion date of the thesis. The Master thesis student must present the results of the thesis research at an announced division-based seminar. The thesis must be approved by the supervisor and two readers.
4.0 PhD Degree Requirements
A research supervisor is assigned for each admitted PhD student.
4.1 PhD Course Work
PhD students are required to complete a minimum of 6 credit hours from 300-level courses (after a Master’s degree). Courses are determined according to the research supervisor. An additional 30 credit hours of course work are required for students bypassing the Master program. They must satisfy similar degree requirements for Master degree without thesis program presented above.
4.2 PhD Qualification
PhD student must pass the qualification exam. The qualification exam consists of two phases. Phase one is the breadth requirements. Each student must demonstrate broad knowledge in the three main areas of Computer Science: systems, theory, and applications. Each of these areas is further subdivided into areas that represent Computer Science fields. Systems include software engineering, operating systems, programming languages, hardware, and software systems. Theory includes algorithms and complexity. Applications include databases, networks, artificial intelligence, and graphics. A PhD student should have already taken a number of advanced courses during graduate and senior undergraduate levels in the above mentioned broad range of categories with a minimum grade of B+ (or equivalent). The student supervisor can determine the breadth course coverage. Based on results of the student transcript and presented course syllabus, each PhD student is given a list of courses to satisfy. The table bellow summarizes the breadth requirement according to the categories and area of courses.
High Performance Computing
282, 291, 292, 311, 312, 380
Systems Hardware and Software Systems
240, 340, 242, 244, 241, 248, 341, 344, 346, 380
Theory Algorithms and Complexity
212, 229, 248, 260, 261, 361
Artificial Intelligence
212, 229, 245, 340, 361, 324
Database
245, 341, 340
Network and Distributed Systems Applications
244, 344, 311, 312
Graphics and User Interfaces
248, 247, 380, 272, 271
The second phase is the qualifying exam. A seminar is required to examine the student knowledge and research skills in a selected topic. PhD student must complete the breadth requirements by the end of the first semester. Once completed, the student should be ready to start working on the PhD qualifying seminar by the end of the first year.
4.3 PhD Candidacy
The PhD Candidacy exam tests the student’s preparedness to pursue thesis research. It is an oral presentation of a research proposal together with questioning by the advisory committee. The student submits a written research proposal to the advisory committee two weeks prior to the exam. The advisory committee consists of a minimum of two faculty-appointed personnel from within the division and the supervisor. The candidate must convince the committee that the chosen research area is suitable and demonstrate an appropriate breadth of knowledge in the chosen area. The committee should decide if there is a thesis topic in the area and whether the candidate is capable of completing such a thesis. The committee decision can be:
- Pass: Student passed and can proceed with the final thesis
- Conditional Pass: Student collects the committee feedback and attempt to complete the deficiencies. The committee can request another informal/ individual oral exam. A Pass must be obtained by the end of the following semester.
- Fail: Student demonstrated unsatisfactory research abilities and is not capable of completing the degree.
The committee reports the results to the student and to the division in writing. Each PhD student must complete the candidacy exam by the end of their forth semester (i.e., second year). PhD students bypassing the Master degree, must complete the candidacy exam by the end of the sixed semester (i.e., third year).
4.4 PhD Seminar Requirement
Each PhD candidate must present at least two publicly announced seminars (or lectures, possibly in 200 level-courses) during the program. The purpose of this requirement is ensures that each student participates in the academic life of the university and to enhance their presentational skills. Each seminar must be attended by the supervisor and one faculty.
4.5 PhD Thesis
The PhD thesis which contains the candidate’s scholarly work. The work of the thesis must be publishable in well recognized journals and conferences.
4.6 PhD Oral Defense
Each student must arrange a two-hour oral defense. The supervisor, advisory committee, external to the division (computer science) examiner, external to the university examiner, and session moderator must be present. The student presents his or her results and answers questions from the examiners. The PhD oral defense result is either Pass or Fail.
COMPUTER SCIENCE PROGRAM COURSES
CS 207. Programming Methodology and Abstractions (3-0-3) Computer programming and the use of abstractions. Software engineering principles of data abstraction and modularity. Object-oriented programming fundamental data structures (such as stacks, queues, sets) and data-directed design. Recursion and recursive data structures (linked lists, trees, graphs). Introduction to basic time and space complexity analysis. The course teaches the mechanics of the C, C++ or Java language. This course is considered remedial training for students in the CS program and will not count toward any degree requirement.
CS 209. Digital Systems (3-0-3) Prerequisites: facility with at least one programming language (at least at the level of CS 207) and logic. The design of processor-based digital systems. Instruction sets, addressing modes, data types.Assembly language programming, low-level data structures, introduction to operating systems and compilers. Processor microarchitecture, microprogramming, pipelining. Memory systems and caches. Input/output, interrupts, buses and DMA. System design implementation alternatives, software/hardware tradeoffs. Labs involve the design of processor subsystems and processor-based embedded systems.
CS 210. Applied Probability and Biostatistics (3-0-3) Prerequisites: Advanced and multivariate calculus. Probability: random variables, independence, and conditional probability; discrete and continuous distributions, moments, distributions of several random variables. Topics in mathematical statistics random sampling, point estimation, confidence intervals, hypothesis testing, nonparametric tests, regression and correlation analyses.Applications in engineering, industrial manufacturing, medicine, biology, and other fields.
CS 211. Numerical Optimization (3-0-3) (Same as AMCS 211.) Recommended prerequisite: AMCS 171. Solution of nonlinear equations. Optimality conditions for smooth optimization problems. Theory and algorithms to solve unconstrained optimization, linear programming, quadratic programming, global optimization, general linearly and nonlinearly constrained optimization problems. Programming project.
CS 212. Linear and Nonlinear Optimization (3-0-3) (Same asAMCS 212.) Optimization theory and modeling. The role of prices, duality, optimality conditions and algorithms in finding and recognizing solutions. Perspectives: problem formulation, analytical theory, computational methods and recent applications in engineering, finance and economics. Theories: finite dimensional derivatives, convexity, optimality, duality and sensitivity. Methods: simplex and interior-point, gradient, Newton and barrier.
CS 221. Artificial Intelligence (3-0-3) Prerequisites: working knowledge of basic discrete mathematics (e.g., sets and functions) and proof techniques, programming ability (at least at the level of CS 207) and exposure to probability. An introduction to the principles and practices of artificial intelligence. Topics include: search, constraint satisfaction, knowledge representation, probabilistic models, machine learning, neural networks, vision, robotics and natural language understanding.
CS 229. Machine Learning (3-0-3) (Same asAMCS 229.) Prerequisites: linear algebra and basic probability and statistics. Familiarity with artificial intelligence recommended. Topics: statistical pattern recognition, linear and non-linear regression, nonparametric methods, exponential family, GLIMs, support vector machines, kernel methods, model/ feature selection, learning theory, VC dimension, clustering, density estimation, EM, dimensionality reduction, ICA, PCA, reinforcement learning and adaptive control, Markov decision processes, approximate dynamic programming and policy search.
CS 240. Operating Systems and Systems Programming (3-0-3) Prerequisite: solid computer programming skills (at least at the level of CS 207). Operating systems design and implementation. Basic structure; synchronization and communication mechanisms; implementation of processes, process management, scheduling and protection; memory organization and management, including virtual memory; I/O device management, secondary storage and file systems.
CS 241. Probability and Random Processes (3-0-3) Prerequisites: Advanced and multivariate calculus. Introduction to probability and random processes. Introduction to probability and random processes. Topics include probability axioms, sigma algebras, random vectors, expectation, probability distributions and densities, Poisson and Wiener processes, stationary processes, autocorrelation, spectral density, effects of filtering, linear least squares estimation and convergence of random sequences.
CS 242. Programming Languages (3-0-3) Prerequisites: programming experience with Lisp, C and an object-oriented programming language. Central concepts in modern programming languages, impact on software development, language design trade-o s and implementation considerations. Functional, imperative and object oriented paradigms. Formal semantic methods and program analysis. Modern type systems, higher-order functions and closures, exceptions and continuations. Modularity, object-oriented languages, and concurrency. Runtime support for language features, interoperability and security issues.
CS 243. Compilers (3-0-3) Prerequisites: solid computer programming skills (at least at the level of CS 207), familiarity with formal languages (regular expressions and grammars). Principles and practices for design and implementation of compilers and interpreters. Topics: lexical analysis, parsing theory, symbol tables, type systems, scope, semantic analysis, intermediate representations, run-time environments, co de generation and basic program analysis and optimization. Students construct a compiler for a simple object-oriented language during course programming projects.
CS 244. Computer Networks (3-0-3) Prerequisite: Operating systems and systems programming. Packet switching, Internet architecture, routing, router architecture, flow control algorithms, retransmission algorithms, congestion control, TCP/IP, detecting and recovering from errors, switching, Ethernet (wired and wireless) and local area networks, physical layers. clocking and synchronization.Assignments introduce network programming, including sockets, designing a router and implementing a transport layer.
CS 245. Databases (3-0-3) Prerequisites: working knowledge of basic discrete mathematics (e.g., sets, functions and relations) and programming skills. Database design and use of database management systems for applications. The relational model, relational algebra and SQL, the standard language for creating, querying and modifying relational and object-relational databases. XMLdata including the query languages XPath and XQuery. UMLdatabase design and relational design principles based on functional dependencies and normal forms. Other topics include indexes, views, transactions, authorization, integrity constraints and triggers.Advanced topics from data warehousing, data mining, Web data management, Datalog, data integration, data streams and continuous queries and data-intensive Web services.
CS 247. Scientific Visualization (3-0-3) Prerequisites: Advanced and multivariate calculus, and linear algebra, computer graphics, and programming experience. Techniques for generating images of various types of experimentally measured, computer generated, or gathered data. Grid structures Scalar field visualization. Vector field visualization. Particle visualization. Graph visualization. Animation.Applications in science, engineering, and medicine. CS 248. Computer Graphics (3-0-3) Prerequisites: solid programming skills and linear algebra. Input and display devices, scan conversion of geometric primitives, 2D and 3D geometric transformations, clipping and windowing, scene modeling and animation, algorithms for visible surface determination, local and global shading models, color and real-time rendering methods.
CS 248. Computer Graphics (3-0-3) Prerequisites: solid programming skills and linear algebra. Input and display devices, scan conversion of geometric primitives, 2D and 3D geometric transformations, clipping and windowing, scene modeling and animation, algorithms for visible surface determination, local and global shading models, color and realtime rendering methods.
CS 251. Numerical Linear Algebra (3-0-3) Prerequisites: Programming skills and linear algebra. Linear algebra in the presence of floating point rounding error, with applications to large-scale scientific computing. Matrix factorizations. Linear least squares.Accuracy and stability. Eigenanlysis. Singular Value Decomposition. Krylov subspace methods and preconditioning. Optimization and saddle-point systems.
CS 260. Design and Analysis of Algorithms (3-0-3) Prerequisites: computer programming skills, knowledge of probability, understanding of basic data structures, basic knowledge in discrete mathematics. Fulfills University Mathematics Requirement. Review of algorithm analysis (search in ordered array, binary insertion sort, merge sort, 2-3 trees, asymptotic notation). Divide and conquer algorithms (master theorem, integer multiplication, matrix multiplication, fast Fourier transform). Graphs (breadth-first search, connected components, topological ordering, depth-first search). Dynamic programming (chain matrix multiplication, shortest paths, edit distance, sequence alignment). Greedy algorithms (binary heaps, Dijkstra’s algorithm, minimum spanning tree, Huffman codes). Randomized algorithms (selection, quick sort, global minimum cut, hushing). P and NP (Cook’s theorem, examples of NP-complete problems). Approximate algorithms for NP-hard problems (set cover, vertex cover, maximum independent set). Partial recursive functions (theorem of Post, Diophantine equations). Computations and undecidable problems (undecidability of halting problem, theorem of Rice, semantic and syntactical properties of programs).
CS 261. Algorithmic Paradigms (3-0-3) Prerequisite: Familiarity with discrete algorithms at the level of AMCS 260. Topics: algorithms for optimization problems such as matching, max-flow, min-cut and load balancing. Using linear programming, emphasis is on LP duality for design and analysis of approximation algorithms.Approximation algorithms for NP-complete problems such as Steiner trees, traveling salesman and scheduling problems. Randomized algorithms.
CS 272 Geometric Modeling (3-0-3) (Same as AMCS 272.) Prerequisites: Advanced and multivariate calculus, and linear algebra, computer graphics, and programming experience. Terminology, coordinate systems, and implicit forms. Parametric and spline representations of curves and surfaces and their uses. Basic differential geometry of curves and surfaces. Subdivision surfaces. Solid modeling paradigms and operations. Robustness and accuracy in geometric computations. Applications.
CS 282. Computer Architecture and Organization (3-0-3) Prerequisite: CS 209. Advanced topics in cache hierarchies, memory systems, storage and IO systems, interconnection networks and message-passing multiprocessor systems (clusters). Issues such as locality, coarse-grain parallelism, synchronization, overlapping communication with computation, hardware/software interfaces, performance/ power trade-offs and reliability. Characteristics of modern processors that affect system architecture.
CS 291 Scientific Software Engineering (3-0-3) Prerequisites: Programming experience and familiarity with basic discrete and numerical algorithms. Practical aspects of application development for high performance computing. Programming language choice; compilers; compiler usage. Build management using make and other tools. Library development and usage. Portability and the GNU autoconf system. Correctness and performance debugging, performance analysis. Group development practices and version control. Use of third-party libraries and software licensing.
CS 292 Parallel Programming Paradigms (3-0-3). Prerequisites: Programming experience and familiarity with basic discrete and numerical algorithms. Distributed and shared memory programming models and frameworks. Thread programming and OpenMP. Message passing and MPI. Parallel GlobalAddress Space (PGAS) languages. Emerging languages for manycore programming. Elements to be covered will include syntax and semantics, performance issues, thread safety and hybrid programming paradigms.
CS 308. Stochastic Methods in Engineering (3-0-3) Prerequisite: CS 241. Review of basic probability; Monte Carlo simulation; statespace models and time series; parameter estimation, prediction and filtering; Markov chains and processes; stochastic control and stochastic differential equations. Examples from various engineering disciplines.
CS 311. High Performance Computing I (3-0-3) Prerequisite: Programming experience and familiarity with basic discrete and numerical algorithms. Part one of a two-course sequence in high performance computing technology, with an emphasis on using KAUST’s research computing systems, focusing primarily on hardware architectures. History of high performance computing. Hardware architectures. CMOS processor design. Cache architectures. Memory architectures. Hardware counters. Processing benchmarks. Power. Single-node performance of real applications.
CS 312. High Performance Computing II (3-0-3) Prerequisites: Programming experience and familiarity with basic discrete and numerical algorithms and AMCS 311. Part one of a two-course sequence in high performance computing technology, with an emphasis on using KAUST’s research computing systems, focusing primarily on hardware architectures. I/O systems and communication networks. Communication benchmarks. Theoretical and achievable performance for processor, memory system, network, and I/O. Future architecture directions and limitations. The course is intended to develop a deep understanding of the underlying high performance computing architectures on which the student will develop and deploy applications.
CS 330. Computational Science and Engineering (3-0-3) Prerequisite: Programming experience and familiarity with basic discrete and numerical algorithms and experience with one or more computational applications. Case studies of representative and prototype applications in partial differential equations and mesh-based methods, particle methods, ray-tracing methods, transactional methods.
CS 337. Information Networks (3-0-3) Prerequisite: probability. Network structure of the Internet and the Web. Modeling, scale-free graphs, small-world phenomenon.Algorithmic implications in searching and interdomain routing, the effect of structure on performance. Game theoretic issues, routing games and network creation games. Security issues, vulnerability and robustness.
CS 340. Computational Methods in Data Mining (3-0-3) Prerequisites: Probability and scientific computing. Focus is on very-large-scale data mining. Topics include computational methods in supervised and unsupervised learning, association mining and collaborative filtering. Individual or group applications oriented programming project. 1 unit without project; 3 units requires final project.
CS 341. Advanced Topics in Data Management (3-0-3) Prerequisites: CS 245 Topics in Data Management will be analyzed and discussed.
Students will engage in research and project presentations. Topics will vary by semester.
CS 344. Advanced Topics in Computer Networks (3-0-3) Prerequisites: CS 244. Solid computer networks background, excellent skills in C/C++ and TCL, suing network simulators such as NS-2, working with Linux systems. Topics in Computer Networks will be analyzed and discussed. Topics will vary by semester.
CS 346. Advanced Topics in Operating Systems (3-0-3) Prerequisites: Solid computer programming skills (at least at the level of CS 207) and solid background in at least one operating systems (CS 240) or computer architecture (at least at the level of CS 209 or CS 282), or permission of instructor. Topics in Operating Systems will be analyzed and discussed. Topics will vary by semester.
CS 380. GPU and GPGPU Programming (3-0-3) Prerequisite: CS 280. Recommended optional prerequisites: CS 248, CS 292.
Architecture and programming of GPUs (Graphics Processing Units). Covers both the traditional use of GPUs for graphics and visualization, as well as their use for general purpose computations (GPGPU). GPU many-core hardware architecture, shading and computer programming languages andAPIs, programming vertex, geometry, and fragment shaders, programming with CUDA, Brook, OpenCL, stream computing, approaches to massively parallel computations, memory subsystems and caches, rasterization, texture mapping, linear algebra computations, alternative and future architectures.
CS 361 Combinatorial Machine Learning (3-0-3) Prerequisites: CS 260. Lower and upper bounds on complexity and algorithms for construction (optimization) of decision trees, decision rules and tests. Decision tables with one-valued decisions and decision tables with many-valued decisions.Approximate decision trees, rules and tests. Global and local approaches to the study of problems over infinite sets of attributes.Applications to discrete optimization, fault diagnosis, pattern recognition, analysis of acyclic programs, data mining and knowledge discovery. Current results of research.
CS 297. Master Dissertation Research (variable credit)
Master-level supervised research.
CS 298. Master Graduate Seminar (variable credit)
Master-level seminar focusing on special topics within the field.
CS 299. Master Directed Research (variable credit)
Master-level supervised research.
CS 397. Doctoral Dissertation Research (variable credit)
Doctoral-level supervised research.
CS 398. Doctoral Graduate Seminar (variable credit)
Doctoral-level seminar focusing on special topics within the field.
CS 399. Doctoral Directed Research (variable credit)
Doctoral-level supervised research.