In this project, we propose a modular continual learning approach to face the problem of catastrophic forgetting and transfer in learning from evolving task distributions. Concretely, we propose a model that learns how to select most relevant modules based on a local decision rule for a given task to form a deep learning model for […]

Chaque jour, des decisions doivent etre prises quant a la quantite d’hydroelectricite produite au Quebec. Ces decisions reposent sur la prevision des apports en eau dans les bassins versants produite a l’aide de modeles hydrologiques. Ces modeles, qui transforment les precipitations en debits, prennent en compte plusieurs facteurs, dont notamment la presence ou l’absence de […]

This research project will create data dashboards that measure Kingston’s economic resiliency a year into the pandemic. The data will be collected from multiple organizations and sources to create a comprehensive look at the pandemic’s impact. The data dashboards will cover the areas of business and tourism, health and community, mobility and home and employment. […]

The representation of the Black, Indigenous, and People of Colour (BIPOC) community decreases at higher levels of education/academia. This phenomenon is proverbially referred to as the ‘leaky pipeline’ and has been studied extensively with various populations in multiple disciplines. However, despite scholars identifying some key factors that contribute to the persistence of the BIPOC community […]

Researchers are often tasked with finding related works in their respective field, including journal entries and images that represent the work contributed, which allows researchers to summarize and expand previous contributions. Biological researchers often convey their contributions through visual illustrations, also considered gene model maps. These gene model maps are used to represent how contributions […]

Ontology induction from unstructured text is a long standing goal of Natural Language Processing (NLP). An ontology can serve a number of tasks, such as question answering, Information Retrieval to name a few. In fact, practice communities are often relying on an ontology in their daily decision making process. This is the case of IATA, […]

With the availability of low-cost edge devices equipped with color and depth sensors, such as iPhone or iPad, 3D data capturing is becoming more accessible and convenient. Asset managers and building owners seek benefiting from this potential to accurately and automatically create 3D indoor models of buildings. In particular, having 3D indoor models containing objects […]

Le monitoring en temps réel des bioprocédés permet d’augmenter les performances et de réduire les pertes. La principale difficulté provient de la quantité limitée d’information accessible en temps réel et des phénomènes complexes et fortement non-linéaires en présence. Ainsi, les algorithmes statistiques conventionnels ne permettent pas d’interpréter adéquatement et avec la précision requise en pharmaceutique […]

Equipment failure is the primary source of unplanned downtime in industries working with rotating electrical machines. Fault detection at the early stages is an essential solution for reducing this downtime. Condition monitoring of machinery is the process of capturing and monitoring parameters such as vibrations to identify a developing fault. This project uses the data […]

This research is motivated by the need to coordinate matching and pricing decisions in resource sharing platforms. In general, matching decisions can be seen from different standpoints of shippers, carriers, and the platform. We are particularly interested in investigating the matching problem considering all standpoints simultaneously, which has not received much attention in the literature […]

Criterium Group will support the Energy Client in transforming operations and determining the most sustainable and economic decarbonization technologies to utilize at The Asset. The Asset site has important characteristics that facilitate bridging to the new technological solutions being pursued as well as beneficial infrastructure, integration, and geological attributes that could improve the feasibility and […]

CARIS is leading a project with the Ocean Supercluster for the development of new innovative cloud technology for delivering CARIS software applications in a Software-as-a-Service (SaaS) model. CARIS tools used worldwide for the processing and management of hydrographic data will be deployed over cloud services, lowering the capital costs through a SaaS subscription business model, […]