Discover current and upcoming courses on AI, ML, and Data Science at Dalhousie University. More on the way!
Course type: Foundational | By Evangelos Milios and Sageev Oore
This course aims to introduce machine learning covering traditional methods to recent deep neural network architectures. The course focuses on the underlying mathematical and statistical principles that underpin those methods and how understanding these principles lead to the successful application of machine learning to address practical problems. Examples of suchproblems will be posed for text and image processing. The following topics will be introduced: linear regression, optimization by gradient descent, support vector machines, decision tree classifiers, naive Bayes classifiers, multilayer neural networks, clustering, evaluation of classification and clustering algorithms, cross-validation, and avoiding overfitting. Students will use a standard machine learning library to apply algorithms to various datasets.
Course type: Foundational | By Evangelos Milios
This course aims to continue the introduction to machine learning building on CSCI3151. The course covers ensemble methods for classification, deep neural networks and their application to structured data, images, and sequences, deep generative models, deep language models, few-shot learning, dimensionality reduction methods, active learning, semi-supervised learning, and Bayesian networks. The course focuses on the underlying mathematical and statistical principles that underpin those methods and how understanding these principles leads to the successful application of machine learning to address practical problems. Students will use standard machine learning libraries to apply algorithms to various datasets.
Course type: Application | By Ga Wu
This course provides a comprehensive exploration of Information Retrieval (IR) systems, ranging from traditional methods to cutting-edge approaches powered by Language Model (LM) technologies. Students will delve into the fundamental principles, techniques, and algorithms underlying search engines, recommender systems, and other IR applications. Through a blend of theoretical lectures, hands-on exercises, and real-world case studies, students will gain a deep understanding of IR concepts and their practical implications.
Course type: Data science | By Frank Rudzicz
This course reviews main concepts in data mining and data warehouses including objectives, architectures, algorithms, implementations, and applications. The topics covered include operational information process, decision-oriented information process, data warehousing, online analytical processing (OLAP), clustering, and classification. Selected system tools for data mining and data warehousing are introduced.
Course type: Data science | By Evangelos Milios and Ga Wu
A recent article in the Harvard Business Review regarded 'Data Scientist' as the most appealing job of the 21st century. There are several reasons for this claim, but the main one comes from the diverse set of skills needed and the shortage of professionals with such background or experience. So, what’s data science about after all? It’s about asking the right questions to transform data into business value using statistics and algorithms. While other fields concentrate on finding previously unknown knowledge or searching for a specific pattern, data science focuses on answering deep questions and making conclusions understandable to the rest of the organization.
Course type: Application | By Finlay Maguire
Health data science is a rapidly growing research field across academia, government, and industry. It relates to the application of statistical and machine learning approaches to analyse large complex medical datasets including electronic medical records, radiological imaging, physiological sensor data, and longitudinal databases. This course combines an overview of these key types of medical data, hands-on introduction to their principal analysis methods, and training in how to apply them in interdisciplinary research contexts. Using a combination of lectures, R-based exercises, student-driven tutorials, and collaborative development of a research proposal, students will gain the skills necessary to plan and conduct effective health data science research.
Course type: Application | By Vlado Keselj
Course type: Application | By Frank Rudzicz
This course introduces spoken language technologies, with an emphasis on deep learning and traditional machine learning for automatic speech recognition, speech synthesis, paralinguistic tasks (e.g., affect detection), and dialogue, with applications to digital assistants and conversational agents. The course is designed to give practical and scientific experience in speech language systems using modern technologies.
Course type: Application | By Carlos Hernandez Castillo
This course introduces students to the fundamental concepts of computer vision providing an overview of the current methodologies and techniques. Students will explore the theory behind fundamental processing tasks, including segmentation, feature extraction, image classification, and object detection, using a mathematical framework to analyze images as two-dimensional signals. By the end of this course, students will be able to apply the basic principles and tools used in computer vision to solve practical problems in scientific and commercial settings.
Course type: Foundational | By Thomas Trappenberg
This course starts with a short introduction to using basic machine learning methods with Python, but the emphasis of this course is the foundation of data modelling in a probabilistic framework. We discuss deep learning, convolution neural networks, generative networks, and transformers, probabilistic regression and bayesian causal networks. While there are many courses that provide recipes on how to use neural networks, the aim of this course is to get a deeper understanding of the scientific principles. This course is specifically recommended for thesis students in machine learning and students that study for a managerial level as opposed to a focus on coding instructions.
Course type: Data science | By Ga Wu
Personalized content recommendation is probably the most widely recognized and successful field of machine learning application in the real world. This course will discuss the concepts behind content recommender systems and how machine learning algorithms could help estimate and track user preference. Topics include a series recommender systems from classic, static, matrix factorization-based system to advanced, dynamic, deep learning-driven systems. Students will gain hands-on experience implementing large-scale recommender systems that meet the standards of real-world applications. They will also learn how to customize and optimize machine learning models for specific tasks by understanding practical constraints in real productions, such as efficiency, scalability requirements.
Course type: Data science | By Evangelos Milios
This course will introduce the concepts of Visual Analytics (VA). VA is a multi-disciplinary domain that combines data visualization with machine learning and other automated techniques to help people make sense of data. Students will be introduced to the design of visual representations supporting tasks to go from findings to insights based on data. Topics include basic concepts of information visualization and machine learning; visual analytics of evolving phenomena; analysis of spatial and temporal data sets; visual social media analytics; and the visual analytics of text and multimedia collections. Students will prototype visual analytics applications using existing toolkits, coupling machine learning and visualization methods. Students will gain competence in performing data analysis and visualization tasks in different application domains.
Course type: Foundational | By Janarthanan Rajendran
Reinforcement Learning (RL), as a branch of machine learning focuses on goal-directed learning from interactions and sequential decision making in complex, dynamic, and uncertain environments. Potential applications of RL are diverse and include applications in areas of robotics, transportation, game playing, healthcare, psychology, neuroscience, industrial process control, and business management, to name a few. This course focuses on RL with function approximation by deep artificial neural networks (Deep RL). Students will learn about the recent advances in the field, including a wide range of deep RL methods, their applicability to different scenarios, their strengths, and their limitations.
Course type: Application | By Hassan Sajjad
Natural Language Processing (NLP) is a field of Artificial Intelligence that enables computers to understand human language. It targets various applications such as, dialog systems like Amazon Echo, translation systems like Google translate, and text generation like auto-completion feature while writing emails. In recent years, neural network based approaches, particularly deep learning, have substantially improved the performance across various NLP tasks and have become the most common approach to target NLP problems. This course aims at providing students with an in-depth knowledge of solving NLP problems with deep learning. It begins with providing an introduction to neural network models and advances to the latest neural network architectures such as RNNs and Transformers. The students get an opportunity to see the application of deep learning models across major NLP problems and familiarize themselves with peculiarities involving text processing.
Course type: Application | By Frank Rudzicz
This course provides a broad overview of machine learning and machine learning operations in healthcare contexts. We begin by studying how healthcare data is unique, and how machine learning methods have been applied to clinical and medical tasks. We focus on various graphical, deep learning, time-series, and transfer learning models and uinique aspects of their application in healthcare. We cover concepts of fairness, privacy, trust, explainability, and other human factors. We discuss imple- mentation techniques, including ‘MLOps’ for healthcare, and opportunities for real-world deployment. Much of the course will be seminar-based, including guest lectures and descriptions of research papers. Students will choose and complete a commensurate research project.