Last updated: August 3, 2022 @ 11:45AM
Master of Science in Computer Science
Faculty of Mathematics and Science
Faculty of Mathematics and Science
Sheridan Houghten (Computer Science), Beatrice Ombuki-Berman (Computer Science), Ke Qiu (Computer Science), Brian Ross (Computer Science), Michael Winter (Computer Science)
Naser Ezzati-Jivan (Computer Science), Robson De Grande (Computer Science), Yifeng Li (Computer Science)
Ivo Düntsch (Computer Science)
Participating Graduate Faculty
Frank Fueten (Earth Sciences), Omar Kihel (Mathematics), Thomas Wolf (Mathematics)
John Bonnett (History), Ping Liang (Biological Sciences)
Joseph Brown, Kyle Harrison, James Hughes
Graduate Program Director
Graduate Administrative Coordinator
905-688-5550, extension 3115
Mackenzie Chown D473
The Department of Computer Science offers a program leading to the Master of Science (MSc) degree. Graduate research topics may be conducted in the broad areas of computational logic and algebra, data mining, evolutionary computation, artificial intelligence, algorithms, parallelism, and combinatorics. Please see the department web page for a listing of faculty and their specific research interests (http://www.cosc.brocku.ca/graduate).
The program offers two options: a thesis option (suitable for students planning doctoral studies or industry employment) and a project option (suitable for students planning industry employment).
Successful completion of four year Bachelor's degree, or equivalent, in Computer Science, with at least a minimum average of 75 (B). In some circumstances, exceptional applicants with four year Bachelor's degree in a related discipline (e.g. mathematics, computer engineering) who have met the minimum average of 75 (B), and have a demonstrated proficiency in fundamental computer science topics (see list below), may be considered. Such applicants may consider submitting a result from the Graduate Record Examination (GRE) subject test in computer science to strengthen their application. Agreement from a faculty advisor to supervise the student is also required for admission to the program.
Applicants are expected to have completed courses in the following areas: computer organization, operating systems, file structures and data management, principles of programming languages, data structures, software analysis and design, formal languages and automata, calculus, linear algebra, statistics and/or probability, discrete mathematics, and additional four upper level (third or fourth year) half courses in other topics in computer science. Candidates lacking sufficient background in the area of the intended Master's degree may be required to complete additional preparatory courses in consultation with their supervisor.
Those applicants holding a three or four year Bachelor's degree and who meet academic requirements of an overall B average may be asked to complete a qualifying term/year to upgrade their application. Completion of a qualifying term/year does not guarantee acceptance into the program.
The Graduate Admissions Committee will review all applications and recommend admission for a limited number of suitable candidates.
Part-time study is available.
The computer science MSc offers two options - a thesis option MSc and project option MSc.
For full-time students in the thesis option, the MSc is normally a six term or two year program. Every MSc candidate must prepare and defend a thesis, which demonstrates a capacity for independent work of high scientific calibre. A supervisory graduate committee will guide the student in all aspects of the student's graduate program. Students normally take four half-credit courses in the first year. Courses are selected in consultation with their assigned supervisor.
Degree requirements for the thesis option must include COSC 5F90 (Thesis) and four 5(alpha)00 or above level COSC half-credits, or three such COSC half-credits and one COSC 4(alpha)00 or above level half credit with the approval of the graduate committee. At most one of the graduate level courses may be a directed reading course. All candidates are required to present seminars on their background research and thesis topics as part of the COSC 5F90 thesis course, attend all the seminars of fellow graduate students and departmental seminars.
For full-time students in the project option, the MSc is normally a four term or 16 month program. Every MSc candidate must take six 5(alpha)00 or above level COSC half-credits, and a 5F99 Project. The course requirements permit one COSC 4(alpha)00 or above credit. All candidates must prepare and submit a COSC 5F99 project, which demonstrates proficiency in applying concepts in computer science in a practical application. The project is typically completed in the final two terms of study.
All MSc students can take one-half non-COSC credit, in consultation with their supervisor. Cross-discipline courses that are suitable as an elective for the project option include (but are not limited to): BIOL 5P06; GEOG 5P25; MATH 5P20, 5P21, 5P35, 5P36, 5P92; and MBAB 5P14, 5P17. The availability of these courses will vary in different years. Students should consider these and other suitable courses in consultation with the supervisor.
A campus-wide fiber optic network links all of the university's academic computing facilities. The department's computers form an integral part of this resource. All faculty and graduate students are provided with an account on the departmental server. Most computers on campus can be accessed from microcomputers in any of the laboratories.
Brock is also a full member of the SHARCnet consortium with access to all its high performance clusters of powerful workstations.
In addition to three servers, the department also maintains several PC based labs and UNIX workstations for teaching and research.
Note that not all courses are offered in every session. Refer to the applicable timetable for details
Students must check to ensure that prerequisites are met. Students may be deregistered, at the request of the instructor, from any course for which prerequisites and/or restrictions have not been met.
The preparation and defence of a thesis demonstrating the candidate's ability for independent and original research.
Practical application of techniques & concepts in computer science under the consultation of an assigned supervisor.
Main concepts, problems and applications related to error-correcting codes. Different classes of codes and their properties. Emphasis on algorithms relating to codes, examination of algorithms for encoding and decoding, together with algorithms that may be used in computer searches for specific classes of codes.
Logic in Computer Science
A thorough introduction to mathematical logic, covering the following topics: propositional and first-order logic; soundness, completeness, and compactness of first-order logic; first-order theories; undecidability and Gödel's incompleteness theorem; and an introduction to other logics such as intuitionistic and modal logics. Furthermore, the course stresses the application of logic to various areas of computer science such as computability, programming languages, program specification and verification.
Universal Algebra for Computer Science
Concepts and constructs of Universal Algebra, such as products, subalgebras, homomorphic images and congruences, term algebras, free algebras, its connections with Logic and Model Theory, decidability issues, lattices and relation algebras, and applications in Computer Science such as Type Theory, Specification, Complexity Theory, Uncertainty Management and others.
Introduction to parallel processing, various parallel computational models including both shared-memory and distributed-memory models, speed-up, cost, design and analysis of parallel algorithms and data structures for a variety of problems in searching, sorting, graph theory, computational geometry, strings, and numerical computation; brief introduction to parallel complexity.
Introduction to Lambda Calculus
Introduction to typed and untyped lambda calculi and their semantics. Syntax of the lambda calculus, conversion, fixed points, reduction, Church-Rosser theorem, representation of recursive functions, lambda models. Category theory, cartesian closed categories and categorical models of lambda calculus.
Principles of Mobile Cloud Computing
Fundamentals supporting Mobile Cloud Computing, including task offloading, connectivity of mobile networks, and remote storage. Study of the basics and recent research advancements on principles of mobile computing, distributed applications and services, Cloud computing and virtualization, management and use of resources offered by Cloud service providers, computation offloading and thin-client computing, and application scenarios and selected use cases.
Software Performance Engineering
Concepts, techniques, and metrics in software performance engineering before, during, and after software development. This covers performance practices throughout the software development life cycle, performance and scalability testing and principles of performance evaluation including instrumentation, profiling, measurement, and benchmarking.
Synthesis of computer programs using evolutionary computation. The study of different representations, including tree, linear, grammatical. Theoretical analyses, including the effects of operators, representations, fitness landscapes. Practical applications in problem solving, decision making, classification, computer vision, design.
Robot Control Architectures
Survey of approaches to control in single and multi-robot systems. Examination and study of different mobile robot control architectures, including deliberative, reactive and hybrid with focus on the issues of resolving the fundamental conflict between thinking and acting, i.e., high-level deliberation and real-time control. Other relevant topics including communication techniques in multirobot systems and safety characteristics of the studied control architectures.
Computer Vision and Visual Computer Learning
Introduction to computer vision and pattern classification. The problems of WHAT and WHERE. The issue of knowledge representation and performance. Knowledge consolidation models. The concept of recursive (i.e. evolutionary) computer learning. Visual learning. Guided learning from infallible and fallible experts. Autonomous learning and experimentation. Analysis of HPC architectures conducive to visual computer learning.
Study of basic concepts of evolutionary algorithms (EAs) from a theoretical and application viewpoint. This includes genetic algorithms, evolutionary strategies, genetic programming, problem representation, genetic operations, overall control, theory of EAs and various examples of important applications. Includes related bio-inspired sub-areas such as swarm intelligence, and evolutionary robotics.
Reading course designed for the individual student and subject to final approval by the department graduate committee. Usually offered by the student's thesis supervisor but may also be offered by other faculty members after consultation with the supervisor.
Non-invasive Data Analysis
Study of data analysis using information from the given data based on the rough set data model. Includes an overview of the data modeling process, principles of probability, non-parametric significance testing, data filtering and discretization, model selection, rule validation, case studies.
Probabilistic Graphical Models and Neural Generative Models
Introduction to the theoretical foundations of generative models and their applications in prediction, knowledge discovery, and creative design. Foundations of probabilistic graphical models (PGMs), learning, and inference in traditional PGMs such as hidden Markov models, mixture models, and latent Dirichlet allocation. Undirected neural generative models (NGMs), including Markov random fields, and variants of Boltzmann machines. Directed NGMs, including Helmholtz machines and deep belief nets, variational autoencoders, and generative adversarial networks.
(also offered as COSC 4P83)
An introduction to fundamental reinforcement learning concepts including multi-armed bandits, Markov decision processes, model-based and model-free methods (such as dynamic programming, Monte Carlo methods, and temporal-difference methods) for learning value and policy functions. Approximation solutions including deep reinforcement learning.
Selected Topics in Computer Science
Various advanced topics in computer science offered by faculty members.