A foundation model (FM) is any model that is trained at scale on a broad dataset and can be adapted (e.g., fine-tuned) to a wide range of downstream tasks; current examples include BERT, CLIP and GPT-3. In this project, we investigate the challenges of building foundational models for drug discovery: capturing multi-modal information, explainability, and […]

Une des tâches les plus exigeantes pour les techniciens en conception et fabrication assistée par ordinateur est la création de modèles anatomiques 3D spécifiques à chaque patient. En effet, les techniciens doivent traiter manuellement les scans volumétriques de la bouche des patients pour extraire les modèles 3D des dents existantes et concevoir les dents manquantes. […]

Recent advancements in deep reinforcement learning (RL) have enabled incredible breakthroughs on a wide variety of problems in which computer systems are required to learn through interacting with the environment with no or minimal human intervention. An example of this is DeepMind’s AlphaGo agent, which taught itself to play Go at a superhuman performance. Deep […]

Fee calculation engine is a tool that is utilized for fee assignment of the investors in financial advisory companies. The assigned fees for the clients can be miscalculated by the advisors. An AI-based monitoring system can reduce the mistakes in the fee calculation by comparing a new assessed fee with the previous assignment. To this […]

iTel has built the largest business internet, voice, and data footprint in Canada. We can create cost-effective network solutions anywhere our clients do business – even in the most remote locations across the country. iTel’s mission is to help businesses do more by connecting workplaces anywhere. The overall project aligns with iTel’s mission because it […]

Synthetic population generation is the process of combining multiple socioeonomic and demographic datasets from various sources and at different granularity, and downscaling them to an individual level. Although it is a fundamental step for many data science tasks, an efficient and standard framework is absent. In this project, we propose a multi-stage framework called SynC […]

The over-arching goal of the project is to explore the use of several recently developed self-supervised image representation learning methods in an attempt to improve performance across several biologically relevant benchmarking tasks at Recursion. At recursion, deep-learning based models are used to generate feature embeddings for our imaging data and these embeddings to generate downstream […]

Honest Empathy Innovations Inc. aims to build sustainable mental healthcare solutions with groundbreaking development and innovation in machine learning. We plan to build a next-generation machine learning platform that Canadians can access for their mental health with this project. Our solution is supported by the Canadian Mental Health Association, NEXT Canada, and Yukon Government. Our […]

Regulatory agencies publish several documents that outline the approval process of drugs. These contain valuable information on a drug’s safety, efficacy, etc. along with the feedback of reviewers from the agencies. Current technologies apply machine learning techniques to extract and categorize the unstructured text found in these documents. However, it does not accurately capture information […]

Surgical Safety Technologies Inc. is expanding upon its existing OR Black Box® platform, which will allow users of the platform to build a personalized, user-created library of surgical videos in the cloud. There are many people and groups around the world who will use this video library to make sure that performance evaluations are fair […]

The project revolves around reading and understanding students solution to various mathematical problems. We wish to analyse the work done by students and the solution to these problems and provide math checking capabilities to various types of problems. Furthermore, this project looks to construct a virtual tutor that can analyse students work and provide feedback […]

Ads keep the internet free. But, to keep them from becoming spam and degrading the user’s online experience, they need to be relevant to them. The traditional way the industry does this is by collecting a lot of information about every user and creating profiles that can be used to target users based on their […]