Machine Intelligence

From autonomous vehicles to hospitals, intelligent machines are everywhere-in smart infrastructure, industrial automation, finance, personalized learning, marketing, medicine, and more.

Machine intelligence is the study, development and application of algorithms that can identify patterns in data and, using these insights, make decisions when confronted with new situations. Engineers trained in machine intelligence work on problems as diverse as finding tumours in medical scans, helping retailers predict which items to stock, and helping self-driving cars to find their way.

EngSci's Machine Intelligence major was launched in 2017 as Canada's first undergraduate program in this field. It provides students with a cutting-edge education in the mathematics, computation, computer hardware, and the software engineering behind artificial intelligence, machine learning, and big data analytics.

The major draws on U of T's outstanding expertise in computer, electrical, industrial and robotics engineering to provide a multidisciplinary curriculum. Courses are taught by faculty members from the Departments of Electrical & Computer Engineering, Industrial Engineering, Computer Science, and the Robotics Group from the University of Toronto Institute for Aerospace Studies.

Topics covered include computing (software engineering), information and intelligence (signals, search, optimization), and algorithms and data analytics (learning, statistical reasoning, decision). The curriculum integrates design thinking and a "whole-systems" approach. Students learn to develop, design and implement learning architectures, feature selection from input data, training strategies, and impact assessment in real world settings. Students also explore the important ethical and societal implications of machine intelligence through case studies and discussion-based learning.

The major's "first principles" approach gives students a deep understanding of the underlying mathematics and modelling. The program also integrates knowledge of computer hardware and software in the design and application of machine intelligence tools.

With the University of Toronto's strong tradition in this field, the recent establishment of the Vector Institute, and the city's vibrant landscape for start-ups and established companies, U of T is an ideal location to launch an exciting career in machine intelligence.

Opportunities for graduates are plentiful in this diverse and rapidly developing field at the intersection of engineering and computer science. Many companies are launching new initiatives and seeking knowledgeable engineers as machine intelligence plays an ever-larger role in industries as diverse as consulting, finance, publishing, law, transportation, and retail.

FAQs

What is the difference between EngSci's Machine Intelligence major and an undergraduate degree in computer science?

While there are some commonalities, engineering offers a unique perspective.

First, engineering students benefit from a systems approach integrating computer hardware and software with mathematics and reasoning. They focus not only on algorithm development, but also on the relationship between machine intelligence and computer architecture and digital signal processing.

Second, graduates excel at problem framing and design thinking. Design thinking is a method for practical and creative problem resolution, and encourages divergent thinking to come up with many solutions, and convergent thinking to identify the best one. Students are taught to properly frame and solve problems in MI, and apply MI tools in areas as diverse as finance, education, advanced manufacturing, healthcare and transportation.

Machine intelligence sits at the intersection of smart and efficient software engineering and deep algorithmic and mathematical thinking. EngSci's emphasis on fundamentals and our Faculty's extraordinary strength in software engineering help students understand the underlying concepts, develop facility with the tools of machine intelligence, and see how these adaptive systems fit into larger engineering designs.

What is the difference between EngSci's Machine Intelligence major and other related EngSci majors?

The Machine Intelligence major teaches students the fundamental concepts that allow graduates to develop and apply machine intelligence in a wide variety of industries and contexts.

The Electrical and Computer Engineering major provides a broad background in both the electrical (applied physics-oriented) and computer (computation-oriented) disciplines and their integration. While there is some opportunity to learn about machine intelligence, the core curriculum covers a broader range of topics, rather than focusing on machine intelligence.

The Robotics Engineering major emphasizes a whole-system perspective for robotics, requiring knowledge of sensors, control, electro-mechanical systems and computer programming. Many robotics engineers are interested in using machine intelligence in robotic systems, but they are less likely to focus on creating the next generation of such tools and techniques.

Finally, graduates from the Engineering Mathematics, Statistics and Finance major apply mathematics, optimization and modelling principles to financial engineering instruments. Machine intelligence is a potential tool in this field, and students in this major may have some exposure to machine intelligence, but not to the same rigor or depth as those in the machine intelligence major.

Is this major the only way to study machine intelligence at U of T?

While EngSci's major provides the most focused and in-depth exposure to this subject at U of T, it is not the only avenue available.

The Department of Computer Science has a world-class program in all areas of artificial intelligence. Students in all U of T Engineering engineering programs, except those in EngSci's Machine Intelligence major, can also take a minor in artificial intelligence engineering. While providing less in-depth exposure to the field, the minor shares some courses with EngSci's major.

How many students from this major do a PEY Co-op after third year?

Over half of students in the Machine Intelligence major have participated in the PEY Co-op Program at companies like Intel, Epson, IBM, ModiFace and Qualcomm.

What are the opportunities for graduates of the Machine Intelligence major?

Opportunities for graduates are plentiful. Every day seems to bring a new start-up or a large company announcing a new initiative in the area of machine intelligence and data science.

Companies and start-ups in diverse areas including consulting, finance, healthcare, retail, transportation and publishing are seeking smart and well-trained engineers to help shape the evolution of these technologies.

Graduates of the major are also well-prepared for studies at top graduate schools and research labs, many of which are quickly expanding their graduate programs in this growing field.

Canada and the Toronto area in particular have a rich MI-ecosystem with many opportunities for entrepreneurial graduates to pave their own way.

Why did the Division of Engineering Science launch a major in machine intelligence?

Engineering Science has a long history of developing educational opportunities for students in emerging and rapidly developing disciplines. Our program has been on the forefront of undergraduate education in areas such as robotics, engineering physics, and biomedical engineering.

The EngSci's program's multidisciplinary foundation curriculum and its rigorous approach to mathematics make it the ideal home for an undergraduate engineering major in machine intelligence. The University of Toronto has unique strengths in the field, given its connections to the newly launched Vector Institute, and has been the launching pad for several start-ups in the area.

Sample Courses

ECE367 Matrix Algebra & Optimization

Students gain a grounding in optimization methods and the matrix algebra upon which they are based.

The first part of the course focuses on fundamental building blocks in linear algebra and their geometric interpretation: matrices, their use to represent data and as linear operators, and the matrix decompositions (such as eigen- and singular-vector decompositions) that reveal structural and geometric insight.

The second part of the course focuses on optimization, both unconstrained and constrained, linear and non-linear, as well as convex and non-convex. Conditions for local and global optimality, first and second-order numerical computational techniques, as well as basic classes of optimization problems are discussed.

Applications from machine learning, signal processing, and statistics are used to illustrate the techniques developed.

ECE368 Probabilistic Reasoning

This course focuses on different classes of probabilistic models and how they are used to deduce actionable information from data.

The course starts by reviewing basic concepts of probability including random variables and first and second-order statistics. Building from this foundation the course covers probabilistic models including vectors (e.g., multivariate Gaussian), temporal (e.g., stationarity and hidden Markov models), and graphical (e.g., factor graphs). On the inference side, topics such as hypothesis testing, marginalization, estimation, and message passing are covered.

Applications of these tools cover diverse data processing domains including machine learning, communications, search, recommendation systems, finance, robotics and navigation.

ECE421 Introduction to Machine Learning

An introduction to the basic theory, fundamental algorithms, and computational toolboxes of machine learning. The focus is on a balanced treatment of the practical and theoretical approaches, along with experience with relevant software packages.

Supervised learning methods covered in the course include: the study of linear models for classification and regression, neural networks and support vector machines. Unsupervised learning methods covered in the course include: principal component analysis, k-means clustering, and Gaussian mixture models. Theoretical topics include: bounds on the generalization error, bias-variance tradeoffs and the Vapnik-Chervonenkis (VC) dimension. Techniques to control overfitting, including regularization and validation, are covered.

MIE324 Introduction to Machine Intelligence

This course introduces students to basic techniques of machine intelligence, and illustrates these through case studies.

Techniques include linear and logistic regressions, support-vector machines, and neural networks, and their use to improve decision making through improved predictions or directly in optimization models.

A significant component of the course is exposure to a state-of-the-art machine-learning software framework with a series of assignments.

In the culminating design project, students work in teams to build a larger-scale machine learning application, and communicate and demonstrate their accomplishments.

ROB311 Artificial Intelligence

Students learn the fundamental principles of artificial intelligence (AI), and explore the subject in rigorous mathematical terms.

Topics include the history and philosophy of AI, search methods in problem solving, knowledge and reasoning, probabilistic reasoning, decision trees, Markov decision processes, natural language processing, and elements of machine learning such as neural-network paradigms.

Did you know...?

Students can further their knowledge in student clubs
like UTMIST, UAIG, and aUToronto.

Find more
student clubs here.

Where this major can take you

EngSci alumni are leaders in industry and research. Meet some of our alumni.

Employers for our recent graduates include Accenture, AMD, DiDi Labs, Google, Intel, Qualcomm, and more.

Recent graduates have attended graduate school at Carnegie Mellon University, ETH Zurich, MIT, UC Berkeley, University of Michigan, U of T, and more.

Draper_2020

Chair of the Machine Intelligence major

Professor Stark Draper (ECE)

In his research, Professor Stark Draper cooks up the math that makes your mobile phone work, makes your computer more energy-efficient, and your personal biometric data more secure.

stark.draper@utoronto.ca

© 2020 Faculty of Applied Science & Engineering