Innovative Projects Realized

Integrated Photonics for Large Scale Quantum Computing

In the past few years, a number of research groups around the world have demonstrated small scale, photonic quantum information processing using approaches like Gaussian Boson Sampling, which can function with the current “Noisy, Intermediate Scale” (NISQ) quantum devices. At the same time, the industry partner and other groups have proposed a variety of large-scale photonic quantum computing architectures. These large-scale architectures have yet to be realized in practice and will require large investments to bring to fruition. In order to maximize the chances of success, it is crucial to explore design options as much as possible in calculations and simulations, before setting out to build large scale hardware. This project will use cutting edge mathematical models and computer simulations to carry out a detailed analysis of the impact of realistic performance, errors and loss levels of photonic components on the overall performance of a proposed large scale quantum computer.

Converge Condo Management – Accounting Analyst Fall 2021

We are looking for an Accounting Analyst who is willing to fill in the gaps and create processes on-the-go that will benefit the whole organization by recommending process improvements over various accounting functions. . The successful candidate will work autonomously but closely with two very experienced CPAs – the president of Converge is a CPA and CA; and one of the directors of Converge is a CPA and CFA and will also work with Converge Board of Directors.

Using AI to help first responders assess skin burns

The broad goal of this project is to create and implement a system which is able of assessing and classifying skin
burns (and other types of skin lesions/wounds) using different state-of-the-art machine learning models and
techniques such as EfficientNets, Reinforcement learning, saliency mappers, CAM, etc… The work that will be
completed during this internship will help first responders to better assess the severity of skin burns. In turns, this will allow for a better management of the patients and, therefore, a more efficient treatment of their burns. Of
course, the insight that will be gathered may also help concretizing similar tools for other skin lesions or wounds.

Enrich

We are doing a series of courses/focus groups to explore the technical and marketing challenges people face when they want to create something new. The project involves asking people what they want to build, helping them understand the framework of the customer discovery process, and working with them to determine the customer needs, solution details, and technical challenges that need to be addressed while creating a solution.

Characterizing the effects of suspended sediment on the smoltification of Atlantic salmon in the Restigouche watershed

In recent decades, the number of wild Atlantic Salmon returning to rivers of Eastern Canada has declined to unprecedented lows. One conservation effort by the Gespe’gewaq Mi’gmaq Resource Council (GMRC) was the construction of various sedimentation basins on Little Main Restigouche near Saint Quentin (NB). These basins are designed to prevent potato fields sediment from entering the Restigouche River watershed after rain events. Fine suspended sediments (SS) can damage salmonid health, notably by impairing their smoltification, an important physiological process that adapts migrating juveniles to life in seawater. We propose to investigate the benefits of trapping SS on salmon smoltification in the Restigouche River watershed (NB). In the lab, we will expose smolts to pulses of fine SS collected from sediment basins and measure biological effects (e.g. smoltification and stress markers). Furthermore, we will measure potentially present agrochemicals (e.g. glyphosate, atrazine) on the collected SS, as they can also be harmful to salmonid health.

Developpement d’un modele de classification probabiliste pour la cartographie du couvert nival dans les bassins versants d’Hydro-Quebec a l’aide des donnees de micro-ondes passives

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 neige au sol. En effet, cette information est primordiale durant la fonte printaniere pour anticiper les apports a venir puisqu’au moins 30% du volume de crue peut provenir de la fonte du couvert nival. Il est donc necessaire pour les previsionnistes de pouvoir suivre l’evolution du couvert de neige de facon quotidienne afin d’ajuster leurs previsions selon le phenomene de fonte. Des methodes pour cartographier la neige au sol sont actuellement utilisees a l’lnstitut de recherche d’Hydro-Ouebec (IREQ), mais elles presentent quelques lacunes. Le but du projet sera d’utiliser des donnees de teledetections en micro-ondes passives, plus particulierement la variable du GTV, a l’aide d’une approche statistique afin de produire des cartes de neige/non-neige.

An Ontologically Controlled Way to Compare The Metallurgical Characteristics of Mining Projects

Cognitive AI software using a metallurgy ontology can use semantic network descriptions of mineral deposits and mines to evaluate a mineral deposit and determine which deposit, anywhere in the world, most closely matches its metallurgical characteristics. After finding a suitable match, the user would be able to read the metallurgical report(s) of the similar deposit(s) to compare with their current understanding of the metallurgical considerations of their project. The final outcome would be to learn from the mistakes and triumphs of others, and view how mines with similar challenges handled their specific processing requirements.

Semantic network descriptions are entered into the system using a user interface that is part of MetMatch. The process involves reading a published paper or NI-43-101 about the deposit and capturing the relevant information in a structured way.

For the system to work as intended, a robust ontology of metallurgical concepts would need to be created, using an existing mineral deposit ontology as a guide.

CO2-less hydrogen gas production via RF (radiofrequency fields) methane pyrolysis

One of the major challenges today is reducing greenhouse gas emissions (GHG) into the atmosphere and the increasing demand in the hydrogen energy sector. Currently, Steam Methane Reforming (SMR) is the industry standard in producing H2. Unfortunately, the H2 produced here is classified as “grey hydrogen” as the reaction between methane and water also produces carbon dioxide (CO2). Methane pyrolysis (MP) offers an alternative approach to H2 production as it decomposes methane molecules into H2 and solid carbon only, making the process significantly cleaner than SMR. In addition to CO2-less H2 production, solid carbon co-production has its market value, making the MP a more economical and greener alternative in H2 production. This project proposes the MP using high electric radiofrequency (RF) to trigger methane decomposition. Acceleware Ltd. aims to commercialize this technology, and this project is centered on proving its feasibility on the laboratory scale.

Composing without forgetting

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 solving a
given task. In this framework we generalization to unseen but related tasks emerge through the composition of
those modules. Additionally, we exploit self-supervised learning to further boost performance through test-time
self-supervised finetuning (active remembering). This is of vital importance for Element AI to provide reusable
solutions that scale with new data, without the need of learning a new model for every problem and improving the
overall performance.

Mycorrhizal fungi associated with Garry Oak in British Columbia

Mycorrhizal fungi provide many benefits to plant health. However, their biodiversity is poorly-studied for Garry oak (Quercus garryana). Garry oak ecosystems are threatened in BC and these fungi could play a critical role in tree growth and survival, particularly with habitat loss and climate change providing additional stressors for this species. The proposed study will collect mycorrhizal root samples, fresh and preserved sporocarps from Garry oak trees in existing habitats to document the fungal species that contribute to ecosystem biodiversity. This will be accomplished by use of DNA barcoding technologies to identify and catalog the mycorrhizal fungi of Garry oaks across its current range. A more complete knowledge of Garry oak mycorrhizal species diversity will assist ecosystem health evaluation and habitat restoration by HAT. This new knowledge base will provide important baseline data for future research and monitoring of this at-risk habitat in BC.

Type 2 Diabetes Management using mobile health technology among South Asian Communities: A Feasibility Study

Gini Health has designed a personalized mHealth and eCounseling intervention for the management of Type 2 diabetes in the South Asian community. For this project, the intern will analyze and interpret secondary, anonymized data that have been collected by the Gini Health team in the implementation of their T2DM management program. Gini Health will benefit by having an experienced post-doctoral fellow leading and overseeing the evaluation team to ensure the research is following best standards. The intern will contribute with previous experience and innovative perspectives on the use of technologies to promote behaviour change in the South Asian community in Canada.

Development and Calibration of DEM Materials Based on Rock Penetration Drillibility Parameters

Distinct Element Models (DEM) are a class of material models that represent the material as a domain of small elastic balls bonded by non-linear elastic springs with defined shear and tensile bond strengths. A part of this internship is a follow up work based on work done by another member of Advanced Drilling Group (ADG) member during his internship which was focused on researching a number of rock analogue materials and measuring 8 drilibility parameters, proposed by Baker-Hughes [2], for them. As a follow up to this work, the measured drillibility parameters are going to be used as a reference for creating DEM rock type materials. The standard procedure recommended in the PFC2d library [7] is going to be modified to be able to match the as many of drillibility parameters possible. These DEM materials can be used later in the rock cutting DEM toolkit [3] that has been already developed by this intern to further study the behavior of different rock types in a cutting phenomenon. l