Study of language disorders, theoretical linguistics, and neuroscience suggests language competence involves two interacting systems, typically dubbed syntax and semantics. However few state-of-the-art deep-learning approaches for natural language processing explicitly model two different systems of representation. While achieving impressive performance on linguistic tasks, they commonly fail to generalize systematically. For example, unlike humans, learning a […]

Office 365 est une plateforme infonuagique de Microsoft qui regroupe plusieurs services très populaires tels que Word, Excel, Outlook, OneDrive, Teams. Sherweb, une entreprise de Sherbrooke, offre un produit appelé Office Protect qui permet détecter des activités anormales dans Office365. Office Protect est complémentaire aux services de surveillance offerts par Microsoft et il est mieux […]

Approximately 70% of data collected by ITOps is not actionable, either due to lack of timeliness or context, and real-time normalized data stored in a data lake is needed to produce actionable data. Also, AIOps tools are often poorly configured and lead to alert fatigue. The benefit to the partner organization is that they will […]

La prédiction est un domaine récent qui connaît de grandes avancées grâce à la démocratisation des connaissances via le web et l’augmentation de la puissance de calcul des ordinateurs modernes. Dans le cadre de ce projet effectué au Mouvement des Caisses Desjardins, ces techniques seront évaluées, documentées et testées vis-à-vis différents cas d’utilisation financiers (gestion […]

To make best use of technology on available big data (e.g., airport parking reservation data), we design and develop an innovative big data science solution for constrained dynamic pricing for airport parking reservations in the proposed research project undertaken by the intern. From scientific point of view, such a solution is novel in the sense […]

Ce projet de recherche est fait en collaboration avec le département de Recherche et Développement de la compagnie Optimum Réassurance Inc. Il a pour but de faire usage des concepts et des outils d’analyse prédictive, de l’apprentissage automatique et statistique pour développer des méthodes de modélisation et de prédiction des réclamations et des comportements des […]

This project aims to train machine learning models that use millimetre-wave radar data to non-invasively detect health incidents, such as overdoses in public bathroom stalls. The partner, Brave Technology Coop, has developed a small, wall-mounted, radar sensing unit on which we want to directly deploy real-time gesture detection software, an instance of cutting-edge “Edge AI.” […]

Competitive team-based digital games like League of Legends, Overwatch and Dota 2 are increasingly popular. These games require intense teamwork but achieving good teamwork is challenging because teams typically comprise of strangers who need to learn to work together in an extremely short amount of time. The match outcome depends on how quickly teams develop […]

The most successful computer vision approaches are based on deep learning architectures, which typically require a large amount of labeled data. This can be impractical or expensive to acquire. Therefore, few-shot learning techniques were proposed to learn new concepts with just one or few annotated examples. However, unsupervised methods such as generative adversarial networks (GANs) […]

The overarching objective of this research project is to design, develop, and test, a reporting automation platform that is able to gather data (in the form of metrics, graphs, tables) from several security software and 3rd party tools, aggregate the data if needed, generate metrics, build reports and upload them to another tool with a […]

Cloud, which is no longer a future trend, has led to a paradigm shift in the design of security services. With so many dynamic features offered by cloud service providers, the assessment of potential threats, cloud vulnerabilities and software security concerns can help in securing cloud customers from inside and outside threat. In this research, […]

In the technology sector, there are so many different computing systems that they need a way to communicate with each other. When different technologies can share information and work together to solve tasks, the benefits of computing are amplified. For example, smart home devices are much more convenient when they all communicate with one platform […]