Courses and Schedule

This course covers data science (DS), a science that combines statistics, data processing, machine learning, scientific inquiry, visualization, business analytics, big data, and big models. The purpose of data science is to address or gain insight into a problem in the real world by the application of computational and statistical techniques.

The course will cover the following subjects: Advanced Data Visualization, Feature engineering, Hypothesis Testing, Feature Selection, outlier detection, Unsupervised Learning, Algorithmic Bias, Virtualization and Distributed learning, Deep Learning (MPLs, CNNs), Computer Vision, Natural language processing, tokenization and transformers, Large Language Model Fine-tuning (SFT), Reinforcement learning (RL, RLHF, RLVR), Vision Language Models (VLMs).

Statistical learning is a discipline that involves extracting information from observed data in order to solve a specific problem. Examples include detecting patterns in signals (images, time series), identifying associations between different variables, or highlighting group structures in a dataset.

Graduate seminar in advanced topics on LLM control.

Séminaire de troisième cycle sur les garanties théoriques pour l'audit post-déploiement.

Senior undergraduate and graduate level course. There has been an unprecedented level of interest and investment in autonomous intelligent systems (AIS), such as artificial intelligence (AI) and robotics. The accelerated development and deployment of AIS, however, brought with it a myriad of social, ethical, and legal problems across the globe. As such, content from the young fields of robot ethics and AI ethics not only have much to offer to complement the technical training engineering students receive, but also are of growing interest to students interested in  AIS such as those enrolled in a minor in AI.  
 

In this course, students will learn fundamental concepts such as human values and ethics, and build on these concepts to explore ethical issues pertaining to AIS through a series of real-world case studies. At the end of the course, students should be able to articulate ethics issues pertaining to AIS, identify stakeholders of an AIS, and what different design and technical approaches exist to help to consider values in the design of an AIS. 

This colloquium is designed for students wishing to become the next generation of responsible robotics practitioners. We will explore the rapidly changing domain together. Along the way, we will learn how robots aren’t simply an AI with a physical embodiment, what the latest findings from the domain of human-robot interaction has to teach us, and what issues are new because of the changing technological landscape.

The course is designed to be engaging and interactive. It includes seminar talks given by guest speakers, lectures by the instructor, and student-led activities. By the end of the course, students will be expected to have an understanding of AI-powered robotics and human-robot interaction, and have a working knowledge of the role that responsible roboticists can play in both the AI and robotics industries.

This course covers a selected set of topics in machine learning and data mining, with an emphasis on understanding the inner workings of the common algorithms. The majority of sections are related to commonly used supervised learning techniques, and to a lesser degree unsupervised methods. This includes fundamentals of algorithms on linear and logistic regression, decision trees, support vector machines, clustering, neural networks, as well as key techniques for feature selection and dimensionality reduction, error estimation and empirical validation.

This is a seminar-style course, where the class as a whole will work together in running the course. In the first few lectures, I will provide an overview of LLMs and highlight the challenges. By the end of the course, you should be able to meaningfully contribute to cutting-edge research in natural language understanding.

Generative flow networks, also known as GFlowNets or simply GFN, are a class of generative machine learning models that perform amortised probabilistic inference and are trained to sample from unnormalised distributions as sequential decision making. This course will cover the fundamental aspects of GFlowNets, starting from a motivation and introduction to the method, and progressing towards more advanced concepts and applications, as well as the connection to other generative models and probabilistic inference methods. The course will combine theory with project work, with a special emphasis on the application of GFlowNets for scientific discovery.

Prediction (The Statistical Trap, Probability, Generalization, Repeatability). The Holdout Method (Stability, Internal Validity). Machine Learning Benchmarks (History of Benchmark Design, Evaluation Failures, Dynamic Benchmarks, Multi-Task Benchmarking, Sensitivity). External Validity, Reproducibility, Robustness (Distribution Shift). Adaptivity and Overfitting (Test Set Reuse). Construct Validity (Context, Scope, Lakatosian Defense). Evaluating Uncertainty (Calibration Error, Prediction Sets, Conformal Inference). Forecast Evaluation (Scoring Rules, Accounting for Uncertainty, Confidence Sequences). Games as Benchmarks (History of Benchmark Games). Evaluating Interventions (Temporal Validity, Compliance, Adjustment). Regulation and Deployment (Program Evaluation, Policy Evaluation, Auditing via Sequential Hypothesis Testing).

System Representation as probabilistic graphical models, inference in graphical models, learning parameters from data.

This is an introductory course on reinforcement learning (RL) and sequential decision-making under uncertainty with an emphasis on understanding the theoretical foundation. We study how dynamic programming methods such as value and policy iteration can be used to solve sequential decision-making problems with known models, and how those approaches can be extended in order to solve reinforcement learning problems, where the model is unknown. Other topics include, but not limited to, function approximation in RL, policy gradient methods, model-based RL, and balancing the exploration-exploitation trade-off. The course will be delivered as a mix of lectures and reading of classical and recent papers assigned to students. As the emphasis is on understanding the foundation, you should expect to go through mathematical detail and proofs. Required background for this course includes being comfortable with probability theory and statistics, calculus, linear algebra, optimization, and (supervised) machine learning.

While graphs are intuitively and naturally represented by vertices and edges, such representations are limited in terms of their analysis, both theoretically and practically (e.g., when implementing graph algorithms). A more powerful approach is yielded by representing them via appropriate matrices (e.g., adjacency, diffusion kernels, or graph Laplacians) that capture intrinsic relations between vertices over the "geometry" represented by the graph structure. Spectral graph theory leverages such matrices, and in particular their spectral and eigendecompositions, to study the properties of graphs and their underlying intrinsic structure. This study leads to surprising and elegant results, not only from a mathematical standpoint, but also in practice with tractable implementations used, e.g., in clustering, visualization, dimensionality reduction, manifold learning, and geometric deep learning. Finally, since nearly any modern data nowadays can be modelled as a graph, either naturally (e.g., social networks) or via appropriate affinity measures, the notions and tools studied in this course provide a powerful framework for capturing and understanding data geometry in general.

This is a graduate-level 4 credit course at UdeM, available also via the ISM. It is suitable for CS, statistics, and applied math students interested in data science and machine learning.

This is a course on representation learning in general and deep learning in particular. Deep learning has recently been responsible for a large number of impressive empirical gains across a wide array of applications including computer vision, natural language processing and speech recognition.