Specialization in Artificial Intelligence

stylized picture of people standing in front of a bank of windows

York University has introduced a Specialization in Artificial Intelligence (AI) in its Master of Science of Computer Science program. Students in this specialization take six graduate courses, of which at least five are within the area of AI, in their first two terms. In addition, students conduct a research project that applies AI to a practical problem under the supervision of faculty members and in collaboration with partners in the private or public sector in their last three terms.

Applications to the Specialization in AI are accepted until March 14, 2018.

Below is a list of faculty members who are part of the Graduate Program in Electrical Engineering and Computer Science at York University. It mentions their areas of research interests within AI (many have other research interests as well) and provides a link to their personal homepage or research group for more information about their research.

Robert Allison

Profile
Vision and intelligent interfaces

Aijun An

Profile
Data mining, machine learning, information retrieval, and AI

Michael Brown

Profile
Computer vision and AI

Natalija Vlajic

Profile
Machine learning in computer security

Marcus Brubaker

Profile
Machine learning, probabilistic methods, computer vision and computational biology

Suprakash Datta

Profile
Machine learning for bioinformatics

Kosta Derpanis

Profile
Computer vision and machine learning

James Elder

Profile
AI and vision

Petros Faloutsos

Profile
AI for computer games and virtual humans

Gerd Grau

Profile
AI-based materials and process development

Michael Jenkin

Profile
Robotics and AI

Hui Jiang

Profile
Machine learning, speech and language processing, and computer vision

Ingo Fruend

Profile
Human vision and AI

Ali Hooshyar

Profile
Smart grid analysis using AI

Zhen Ming (Jack) Jiang

Profile
Software analytics and software performance engineering

Richard Wildes

Profile
AI and vision

Hossein Kassiri

Profile
AI-based algorithms for decoding a physiological/neurological function

Matthew Kyan

Profile
Machine intelligence approach for virtual environments

Yves Lesperance

Profile
Knowledge representation and reasoning, autonomous agents and multi-agent systems, and cognitive robotics

Peter Lian

Profile
Circuits and systems for embedded AI and neuromorphic computing

Marin Litoiu

Profile
Adaptive software and autonomic computing

Sebastian Magierowski

Profile
Hardware acceleration for machine learning

Manos Papagelis

Profile
Data mining, graph mining, machine learning, big data analytics, knowledge discovery

Ali Sadeghi-Naini

no link to personal webpage yet

AI in precision medicine; Machine learning in image-guided therapeutics; Quantitative imaging and radiomics

Mikhail Soutchanski

Profile
AI planning, planning in hybrid systems, knowledge representation including causality, reinforcement learning for planning

Zbigniew Stachniak

Profile
Automated reasoning and propositional satisfiability

Vassilios Tzerpos

Profile
Machine learning and audio

Ruth Urner

Profile
Machine learning theory

Franck van Breugel

Profile
Reinforcement learning for finding bugs in software

Students have to complete three courses of the following list. These courses are all offered every year.

Courses

EECS 5326 3.0 Artificial Intelligence

This course will be an in-depth treatment of one or more specific topics within the field of artificial intelligence.
Prerequisites: none.

EECS 5327 3.0 Introduction to Machine Learning and Pattern Recognition

Machine learning is the study of algorithms that learn how to perform a task from prior experience. This course introduces the student to machine learning concepts and techniques applied to a pattern recognition problem in a diversity of application areas.
Prerequisites: none.

EECS 6127 3.0 Machine Learning Theory

This course takes a foundational perspective on machine learning and covers some of its underlying mathematical principles. Topics range from well-established results in learning theory to current research challenges. We start with introducing a formal framework, and then introduce and analyze learning methods, such as Nearest Neighbors, Boosting, SVMs and Neural Networks. Finally, students present and discuss recent research papers.
Prerequisites: none.

EECS 6327 3.0 Probabilistic Models & Machine Learning

Intelligent systems must make effective judgments in the face of uncertainty. This requires probabilistic models to represent complex relationships between random variables (learning) as well as algorithms that produce good estimates and decisions based on these models (inference). This course explores both probabilistic learning and inference, in a range of application areas.
Prerequisites: none.

EECS 6412 3.0 Data Mining

This course introduces fundamental concepts of data mining. It presents various data mining technologies, algorithms and applications. Topics include association rule mining, classification models, sequential pattern mining and clustering.
Prerequisites: an introductory course on database systems and an introductory course on probability.

Other AI-related Courses

Students have to complete two other courses from the following list. Only some of these courses are offered (varies year by year).

EECS 5323 3.0 Computer Vision

This course will introduce the basic concepts in Computer Vision. Primarily a survey of current computational methods, we will begin by examining methods for measuring visual data (image based operators, edge detection, feature extraction), and low-level processes for feature aggregation (optic flow, segmentation, correspondence). Finally, we will consider some issues in “high-level” vision systems.
Prerequisites: none.

EECS 5324 3.0 An Introduction to Robotics

This course will introduce concepts in Robotics. The course will begin with a study of the mechanics of manipulators and robot platforms. Trajectory and course planning, environmental layout and sensing will be discussed. Finally, high-level concerns will be introduced. The need for real-time response and dynamic-scene analysis will be covered, and recent developments in robotics systems from an Artificial Intelligence viewpoint will be discussed.
Prerequisites: none.

EECS 5326 3.0 Artificial Intelligence

This course will be an in-depth treatment of one or more specific topics within the field of artificial intelligence.
Prerequisites: none.

EECS 5327 3.0 Introduction to Machine Learning and Pattern Recognition

Machine learning is the study of algorithms that learn how to perform a task from prior experience. This course introduces the student to machine learning concepts and techniques applied to a pattern recognition problem in a diversity of application areas.
Prerequisites: none.

EECS 6127 3.0 Machine Learning Theory

This course takes a foundational perspective on machine learning and covers some of its underlying mathematical principles. Topics range from well-established results in learning theory to current research challenges. We start with introducing a formal framework, and then introduce and analyze learning methods, such as Nearest Neighbors, Boosting, SVMs and Neural Networks. Finally, students present and discuss recent research papers.
Prerequisites: none.

EECS 6322 3.0 Neural Networks and Deep Learning

This course covers the theory and practice of neural networks. Topics covered include training methods and loss functions, automatic differentiation and backpropagation, network architectures for a range of problems (images, text, audio, graphs, etc), validation and model selection, software tools and frameworks for deep learning.
Prerequisites: a foundational course in machine learning including, but not limited to, EECS 5327 or EECS 6327

EECS 6323 3.0 Advanced Topics in Computer Vision

An advanced topics course in computer vision which covers selected topics in greater depth. Topics covered will vary from year to year depending on the interests of the class and instructor. Possible topics include: stereo vision, visual motion, computer audition, fast image processing algorithms, vision based mobile robots and active vision sensors, and object recognition.
Prerequisites: EECS 5323.

EECS 6325 3.0 Mobile Robot Path Planning

The focus of this course is on robot motion planning in known and unknown environments. Both theoretical (computational-geometric) models, as well as practical case studies will be covered in the course.
Prerequisites: none.

EECS 6327 3.0 Probabilistic Models & Machine Learning

Intelligent systems must make effective judgments in the face of uncertainty. This requires probabilistic models to represent complex relationships between random variables (learning) as well as algorithms that produce good estimates and decisions based on these models (inference). This course explores both probabilistic learning and inference, in a range of application areas.
Prerequisites: none.

EECS 6328 3.0 Speech and Language Processing

Introducing the latest technologies in speech and language processing, including speech and recognition and understanding, keyword spotting, spoken language processing, speaker identification and verification, statistical machine translation, information retrieval, and other interesting topics.
Prerequisites: none.

EECS 6332 3.0 Statistical Visual Motion Analysis

A seminar course that examines statistical approaches to visual motion analysis, including 3-D structure and motion estimation, optical flow, segmentation and tracking using tools like Maximum Likelihood Estimation, Maximum A Posteriori, Least Squares and Expectation Maximization.
Prerequisites: none.

EECS 6333 3.0 Multiple View Image Understanding

This course considers how multiple images of a scene, as captured by multiple stationary cameras, single moving cameras or their combination, can be used to recover information about the viewed scene (e.g., three-dimensional layout, camera and/or scene movement). Theoretical and practical issues of calibration, correspondence/matching and interpretation will be considered.
Prerequisites: EECS 5323.

EECS 6340 3.0 Embodied Intelligence

This course is intended as a follow-on from a first course on Artificial Intelligence. Whereas such first courses focus on the important foundations of AI, such a Knowledge Representation or Reasoning, this course will examine how these separate foundational elements can be integrated into real systems. This will be accomplished by detailing some general overall concepts that form the basis of intelligent systems in the real world, and then presenting a number of in-depth cases studies of a variety of systems from several applications domains. The embodiment of intelligence may be in a physical system (such as a robot) or a software system (such as in game-playing) but in both cases, the goal is to interact with, and solve a problem in, the real world.
Prerequisites: introductory courses in artificial intelligence. robotics, and computer vision.

EECS 6390A 3.0 Knowledge Representation

This course examines some of the techniques used to represent knowledge in artificial intelligence, and the associated methods of automated reasoning. The emphasis will be on the compromises involved in providing a useful but tractable representation and reasoning service to a knowledge-based system. The topics may include: formal models of knowledge and belief, systems of limited reasoning, languages of limited expressive power, defaults and exceptions, meta-level representation and reasoning, reasoning about action, and theories of rational agency.
Prerequisites: an introductory course on first-order logic.

EECS 6390D 3.0 Computational Models of Visual Perception

This course examines the problem of developing rigorous computational models for visual processing. Computational strategies may draw upon techniques in statistical inference, signal processing, optimization theory, graph theory and distributed computation.
Prerequisites: none.

EECS 6412 3.0 Data Mining

This course introduces fundamental concepts of data mining. It presents various data mining technologies, algorithms and applications. Topics include association rule mining, classification models, sequential pattern mining and clustering.
Prerequisites: an introductory course on database systems and an introductory course on probability.

EECS 6414 3.0 Data Analytics and Visualization

Data analytics and visualization is an emerging discipline of immense importance to any data-driven organization. This is a project-focused course that provides students with knowledge on tools for data mining and visualization and practical experience working with data mining and machine learning algorithms for analysis of very large amounts of data. It also focuses on methods and models for efficient communication of data results through data visualization.
Prerequisites: introductory courses in algorithms, probability theory, linear algebra, and programming.

AI

Degree Requirements

  • Three courses from the following list: EECS 5326, EECS 5327, EECS 6127, EECS 6327, EECS 6412
  • Two other courses from the following list: EECS 5323, EECS 5324, EECS 5326, EECS 5327, EECS 5326, EECS 6127, EECS 6322, EECS 6323, EECS 6325, EECS 6327, EECS 6328, EECS 6332, EECS 6333, EECS 6340, EECS 6390A, EECS 6390D, EECS 6412, EECS 6414

Other Requirements

  • One other graduate course
  • A research project that applies AI to a practical problem under the supervision of faculty members and in collaboration with partners in the private or public sector
  • At least one course must be from each of the following three areas:
    • Theory of Computing & Scientific Computing (second digit is 1 or 2)
    • Artificial Intelligence & Interactive Systems (second digit is 3)
    • Systems: Hardware & Software (second digit is 4 or 5)
  • No more than two courses can be integrated with undergraduate courses (first digit is 5)
  • An honors degree in Computer Science or equivalent, with at least a B+ average in the last two years of study.
  • The equivalent of a senior-level course in the area of theoretical computer science.
  • Minimum English language test scores (if required): TOEFL 577 (paper-based) or 90-91 (Internet-based), IELTS 7, or York English Language Test 4.
  • The Graduate Record Examination general test is strongly recommended, especially for applicants who did their work outside of Canada and/or the United States.
Students admitted to the Specialization in AI are in the position to apply for scholarships from the Vector Institute (details will be available at a later date). No further financial support will be provided.
photo of the Toronto skyline at night seen from the harbour

York University is just a subway ride from downtown Toronto. With more than 50,000 students, it is the third largest university in Canada. Its student population represents the diversity that characterizes Toronto.

Toronto is one of the most culturally diverse cities in the world, with over 200 distinct ethnic origins represented among its inhabitants. There are over 160 different languages spoken in the city. With almost three million residents, it is the largest city in Canada.

The Lassonde School of Engineering is one of York’s eleven Faculties. The School was established in 2011 and consists of four departments, with the Department of Electrical Engineering and Computer Science as it's largest department. It is housed in the Lassonde Building.

The Graduate Program in Electrical Engineering and Computer Science was established in 1989. It offers three degree programs: Master of Science in Computer Science, Master of Applied Science in Electrical and Computer Engineering, and Doctor of Philosophy in Electrical Engineering and Computer Science. The Specialization in AI, which is part of the Master of Science in Computer Science program, was introduced in 2018.

For more information about the Specialization in AI, please contact gpa@eecs.yorku.ca