Projets novateurs réalisés

Analytic tools for assessing the predictability and trustworthiness of social AI systems

As Artificial Intelligence (AI) becomes more developed and integrated into society, it will no longer consist of single isolated systems, but networks of AIs interacting with humans in socio-technical systems interwoven into the functioning of our society. When systems have many interacting components, they start to interact in unexpected ways and become hard to predict — similar to weather systems, where temperature can have many effects on the atmosphere, oceans, ice caps, land surface, and living organisms which interact in web in unpredictable ways. This complexity and unpredictability can be made worse when a system’s components are interacting, not just physically as in the weather system, but socially with signals and gestures, which can often be misinterpreted and lead to confusion. Therefore, as we make AI more socially intelligent and interconnected, it is important to assess how this changes their predictability; specifically, when do AI systems become unpredictable? In this research, we will explore the dynamics of noise in systems. We will use evolutionary game theory to determine social conditions or tipping points. This will assist in evaluating the trustworthiness of AI systems with this form of social intelligence.

Modelling and characterization of periodically patterned optical ring resonators

Optical gyroscopes are instruments capable of measuring rotations. The creation of miniature optical gyroscopes could lead to the development of self-navigating small crafts such as drones. Microresonator-based optical gyroscopes, where light is trapped in a ring-like structure and propagates differently depending on whether it is moving clockwise or counter-clockwise, are the best candidates to create such miniaturized devices. The detection limit (the slowest rotation they can measure) of resonator-based gyroscopes depends on the losses of the resonator, quantified by a number called the Q-factor. When the losses are low, corresponding to a high Q-factor, the detection limit becomes smaller. To make these gyroscopes able to detect very slow rotations, it is important to maximize the Q-factor. One way to achieve this goal is by making a ring with a grating-like pattern. This project involves computer modelling and theoretical analysis of integrated photonic ring resonators including grating patterns with the goal of predicting where the best optical resonances are and to minimize their losses. In addition, there is an experimental component of the project which is the implementation of experimental techniques that can measure resonant modes with ultra-high Q-factors in silicon nitride ring resonators.

Cross-functional teams to develop and commercialize AI-powered solutions – Part G

THIS IS A GENERIC TEXT PUT IN PLACE AS THERE WAS NO PROJECT OVERVIEW

Diagnostic des communications – Municipalité d’Eastman

La municipalité d’Eastman a développé au fil des ans une multitude de canaux de communication pour joindre sa population, et ce, sans plan formel de communication. Dans une perspective de rationalisation, la municipalité souhaite revoir ses pratiques info-communicationnelles afin d’optimiser le choix de canal et la fréquence des communications en fonction des différents types de contenus à communiquer et de ses contraintes budgétaires.

L’objectif de l’intervention est d’effectuer l’évaluation des pratiques actuelles de communication de la municipalité d’Eastman et d’élaborer un plan de communication préliminaire pour celle-ci. Cet objectif pourra être atteint par :
1. Le diagnostic des pratiques actuelles de communication (canaux, fréquence, contenu, ressources requises, appréciation des citoyens) incluant l’intégration des données existantes extraites du sondage mené par le comité arts et culture;
2. L’identification des meilleures pratiques de communication en milieu municipal;
3. L’analyse comparative des coûts entre les différentes pratiques info-communicationnelles (incluant les dépenses liées aux salaires pour la rédaction du contenu);
4. L’élaboration d’un plan de communication préliminaire.

Les réseaux thermiques comme stratégie de transition énergétique urbaine et de décarbonation – le potentiel de l’hydrothermie

Dans le cadre de ce stage, nous renseignerons le potentiel porté par l’hydrothermie fluviale comme source d’énergie. Ce potentiel sera étudié en considérant le territoire montréalais comme zone d’étude et dans un contexte de transition énergétique et de lutte au changement climatique. Nous suivrons une approche méthodologique et scientifique et développerons un argumentaire visant à promouvoir la solution après avoir identifier les facteurs facilitant et contraignant de l’hydrothermie fluviale.

Process development and pilot scale validation of the commercial scale rubber recycling process using an efficient novel devulcanizing agent

The “Be Tire Smart” campaign of the Canadian government and The Rubber Association of Canada strongly encourages the four R (reduce, reuse, recycle, and recover) principles to deal with the scrap tire problem in Canada. The constantly developing automotive industry and the growing number of vehicles generate massive tire demand while the management and recycling of tires at the end-of-life pose critical environmental and socioeconomic issues. The development of a low-cost, environmentally benign, and industrial-scale rubber recycling methods are gaining significant attention.
The proposed initiative, a partnership between Arduro Sustainable Rubber and the Verschuren Centre focuses on the development of a standard commercial-scale process for rubber recycling. Arduro’s patented non-hazardous chemical process uses an oxidizing agent named ArduroDevulc to break the crosslinks in sulfur vulcanized rubber. Arduro has demonstrated feasibility of performing direct devulcanization of sulfur vulcanized rubber by ArduroDevulc naturally in a desired phase at standard temperatures and pressures; thus, addressing the challenges of using aqueous solution of ArduroDevulc.
The aim of the proposed research project is to make the patented devulcanizing process of Arduro more competitive and cost-efficient before its implementation in the intended commercial-scale facility in Dartmouth, NS, Canada….

Génération de modèle 3D par IA au sein d’une PaaS de rendu 3D temps réel.

Chez 3dverse, nous sommes persuadés que débloquer le potentiel de la 3D en temps réel est la clé pour faire avancer l’innovation. C’est pourquoi notre équipe construit la première véritable plateforme de rendu 3D en temps réel (PaaS). Nous savons que cette technologie aura un impact décisif et changera la façon dont les gens visualisent, interagissent et créent avec la 3D. La demande d’expériences de plus en plus immersives et accessibles croît de manière exponentielle. Il n’est pas surprenant que l’adoption généralisée des expériences et des technologies en 3D soit un changement économique et sociétal majeur. Notre plateforme fournit un moteur 3D à la demande, permettant aux développeurs de construire des expériences 3D de manière transparente, instantanément accessibles par tous les utilisateurs sur n’importe quel appareil, à tout moment. 3dverse élimine ainsi les barrières d’accès à la fabrication et à la distribution de contenu 3D en utilisant des technologies d’informatique en nuage, des réseaux haut débit accessibles et d’Intelligence Artificielle.

Adaptation of a training program for healthcare practitioners to improve their competency in promoting health behaviour change for Colombia

Non-communicable diseases account for 70% of all deaths worldwide and are primarily attributed to health risk behaviours, including insufficient physical activity, low quality diets, and smoking. Despite widespread awareness of these, 43% of Colombians lead sedentary lifestyles, 28% fail to eat enough fruits and vegetables, and 12% smoke, underscoring the need for interventions aimed at fostering behavioural changes and healthier habits. Healthcare practitioners (HCPs) are a key group of people who can help individuals engage in the behaviour change process; however, most HCPs don’t get training in being able to effectively help their patients.

The main objectives of this project are to: translate and culturally adapt an effective training program (MOTIVATOR), which enhances HCPs skills in empowering behaviour change in their patients, to the Colombian context; and then conduct initial validation testing of the capacity of the Colombian version of the program to improve HCPs motivational communication competency. This will be achieved through developing the necessary content and then conducting proof-of-concept testing in a group of Colombian healthcare practitioners.

The expected outcome of this project is the development of an adapted version of the MOTIVATOR program to instruct HCPs in Colombia in motivational communication and behaviour change counseling techniques.

Beyond Numbers: Innovations in Aurora Heat’s Finance and Social Impact Strategy

Aurora Heat, Inc., led by Indigenous entrepreneur Brenda Dragon and headquartered in Fort Smith, NWT, creates sustainable, reusable hand, foot and body warmers from sheared beaver fur. Their mission is to empower Indigenous women and youth through inclusive employment, fair wages, and cultural heritage preservation, all while meeting warmth and health needs for North Americans.

To realize a five-year goal of $7 million in annual revenue and become a top employer for Indigenous people in Canada, Aurora Heat must tackle significant challenges. First, they must enhance resilience to climate change, which disproportionately affects businesses reliant on natural resources. Secondly, they need to innovate their bookkeeping practices to optimize financial management. Finally, Aurora Heat must develop an extensive valuation model and cash flow projections to enhance risk management and guide strategic decisions.

The “Beyond Numbers: Innovations in Aurora Heat’s Finance and Social Impact Strategy” project lays out a strategic approach to these challenges and opportunities. Three key objectives drive this initiative, which include creating a comprehensive valuation model and risk assessment, streamlining bookkeeping procedures, and developing an annual social impact report aligning with global sustainability standards. The project’s intern will play an important role in realizing these objectives by preparing three key deliverables: an extensive valuation model, a bookkeeping innovation report, and an annual social impact report.

Development of dual dynamic covalent network materials

Triboelectric nanogenerators (TENGs) are a new type of energy harvesting device converting mechanical stimuli, such as vibration, human motion, wind, and water waves, to electrical signals through the combination of triboelectric effect with electrostatic induction. However, conventionally-designed TENGs have major challenge that includes the failure of their self-charging power and the suppression of their output performance and lifetime, because of their long-term mechanical impacts and thus susceptible to fracture and surface damage. This proposed research aims to overcome the challenge with the integration of dual dynamic covalent adaptive networks (CANs) in TENG devices, thus developing self-healable and reprocessable TENG devices. We will study self-healing mechanisms in dual modes for dual dynamic CANs crosslinked with two dynamic covalent linkages as well as the enhancement of dielectric properties for the development of robust dual dynamic TENGs. Eventually, our proposed research will provide design principles for energy harvesting SH-TENGs in the development of flexible electronics requiring self-healability and reprocessability.

Développer des modèles pour la détection d’évènements de bas-niveau basés sur les textures des codes haptiques

D-BOX conçoit et fabrique des systèmes de mouvement haptique pour le cinéma maison et le cinéma en salle qui offrent une expérience immersive en générant des effets de mouvements et vibrations synchronisés avec l’action à l’écran. Les formes d’ondes d’effets haptiques utilisées pour activer les sièges sont actuellement générées manuellement à travers un processus fastidieux et complexe. La conception des effets haptiques se résume en deux étapes principales, premièrement à trouver des événements d’intérêt dans les films à supporter par les effets haptiques (détection de l’action), et en deuxième temps à générer des formes d’ondes haptiques qui correspondent aux effets voulus. Ce projet se concentre principalement sur la partie détection des actions. Le but est de développer des modèles d’apprentissage profond (Deep Learning) utilisant le son et le visuel dans les films pour détecter une série d’évènements et leurs chronologie (time-code) qui peuvent être des évènements de haut-niveau comme des coups de feu, explosions, vrombissements de moteur; ou bien des événements de bas-niveau comme ‘vibrations’, ‘mouvements’, ‘secousses’, etc. L’approche consiste à utiliser une topologie à deux flux, l’un étant basé sur le son et l’autre sur la vidéo en tant qu’entrées. Le son sera traité sous forme de spectrogrammes, ce qui permettra de traiter les deux contenus comme des images et ainsi exploiter les deux entrées pour comprendre l’action dans les films.

From Deregulation to Decentralization: The Prospect of Global Governance in Web3

THIS IS A GENERIC TEXT PUT IN PLACE AS THERE WAS NO PROJECT OVERVIEW