Innovative Projects Realized

Deep Learning/Computer Vision for Robotic Manipulation

Research is rapidly progressing in enhancing the Artificial Intelligence of Robotics. One backbone of this rapid change lies in Deep Learning. Deep Learning refers to new algorithms that are capable of learning behaviors after being trained by several thousands or even millions of examples of what should be done given an input. My project will be doing computer vision related research in this field. Computer vision mainly refers to tasks where a significant part of input is from a visual source, like a camera. For example, a robot would receive a video feed from a camera as input, akin to how a human would see from their eyes. My research is to delve into new ways to train a robot such that it can perform human like tasks (such as picking up objects or inserting pegs in holes) primarily using video feed input from a camera. TO BE CONT’D

Audience Allocation to Retail Geo-clusters

Based on the user’s geo-location, timestamp and other attributes (eg. time of day, past visit history and app behavior categories, etc.), a machine learning algorithm can be developed to find which cluster the users belong to. Overall, the data of geo-location and timestamp are used to roughly locate the potential clusters. This project will involve some techniques and algorithms like cloud computing i.e Google Cloud Dataproc, sliding windows, histogram and machine learning algorithms. The challenge of first phase would be coming up with a good way of estimating the number of clusters. Then by applying all the above techniques, the decisive attributes can be decided and combined to determine which cluster the users belong to.

Real-time object recognition on wearable devices

The goal of the project is to implement real-time state of the art object recognition models on wearable devices. These devices aim to help people living with a visual disability by providing a description of their outdoor environment and offer navigation guidance. This would improve the experience of the users by allowing them to perform usual day-to-day tasks with much more ease and safety.

Assessing and Addressing Health Disparities Related to Utilization of Preventive Care Services in Ontario

Health disparities arise as a result of long-standing societal disadvantage and discrimination. As machine learning models become more popular in the healthcare sector, understanding of current health disparities becomes even more critical. Without careful management of existing biases, the models can inherit and amplify health disparities, leading to highly undesirable clinical outcomes. This project focuses on health disparities in access to preventive care services. Preventive care services such as screening and preventive medicine allows for early diagnosis and timely interventions. This project aims to provide an understanding of if and how patterns of preventive care utilization aggravates health disparities in Ontario, by employing advanced data exploration and visualization techniques. After establishing such a relationship, this project also provides an individual risk profiling tool to assess the efficacy of preventive services, using advanced feature representation and deep learning techniques.

Electrical Load Forecasting

Load forecasting is an essential activity for a company like Hydro-Québec. It is necessary for objectives as varied as the management of production or the management and maintenance of the electricity network. Any significant forecasting error can result in reliability issues, loss of opportunity, or additional costs to the business. On the other hand, a good prediction would allow Hydro-Québec to generate additional sales in neighboring markets. With the deployment of its Advanced Measurement Infrastructure (AMI), Hydro-Québec now has a significant amount of new consumption data. This data can be used to improve demand forecasting, increasing reliability, decreasing expenses, and potentially generating new revenue.TO BE CONT’D

Role of transthoracic impedance and current in synchronized electricalcardioversion

Synchronized cardioversion is a medical treatment that applied an electrical pulse to restore a normal heart rhythm is patients with an abnormally fast heart rate or cardiac arrhythmia. A successful cardioversion is dependent on the amount of electrical current that reaches the heart, which depends on the strength of the electrical pulse and the transthoracic impedance (electrical impedance of the body). If a cardioversion is not successful, additional attempts

are often made; however, repeated delivery of a large electrical pulse is not desired. Increasing the strength of the electrical pulse is also not ideal as it increases the chances of complications. This research will investigate the role of transthoracic impedance on cardioversion, which will also include the effect of different paddle placement. Outcomes of this research will help improve the efficacy of cardioversion, increase the success of treatments, while minimizing the strength of the electrical pulse.

Understanding Real-time Particle Systems for Health, Entertainment and VR

The proposed research is a collaboration between Persistant Studios’ PopcornFX and SFU’s iVizLab to collaboratively work on ways to understand the processes involved in content creation using a real-time particle system. The iVizLab’s research focuses on using real-time visuals with the biodata from the users as one of the main interfaces to create affective systems that can intelligently interact with the users. In creating the visuals for the iVizLab, it is important to be able to create content that can be modified in real-time with the incoming data. The intern will be working closely with the partner organization to work on understanding complex design processes and to break them down into simpler components to better understand the involved processes. Further, the intern will be working on ways to document these processes and share this with the community of PopcornFX users in the industry and world-wide.

Chaire de recherche industrielle dans les collèges du CRSNG en fabrication de composantes aérospatiales en matériaux composites

Ce projet vise à répondre aux besoins des industriels canadiens aérospatiaux d’aujourd’hui, spécialisés dans la fabrication de composantes en matériaux composites. Ces besoins ont été définis avec plusieurs PME et donneurs d’ordres, dont Bombardier Aérostructures et Services d’Ingénierie, Hutchinson Aéronautique et Industrie Canada, SphèreCo, Texonic, Lubricor, Génik et PCM Innovation. Ce projet se concentrera sur le développement de procédés de fabrication et de matériaux efficaces, afin d’augmenter la productivité, renforcer la compétitivité de l’industrie locale à l’échelle mondiale et ainsi rapatrier la fabrication de pièces composites au Canada. Dans un deuxième temps, ce projet permettra de combler le besoin criant en personnel hautement qualifié auquel l’industrie des composites fait face.
Effectuée partenariat avec le CTA, ce projet permettra également d’établir une synergie efficace et pérenne entre la recherche universitaire et collégiale.

Simplification of long sentences

The task of sentence simplification can present itself in multiple forms. It could consist in correcting the punctuation of a sentence like so:
Avant : J’ai acheté un bateau je l’aime beaucoup.
Après : J’ai acheté un bateau. Je l’aime beaucoup.
However, a sentence can be both long and written correctly. In this case, it would require a reformulation in multiple sentences like so:
Avant: J’ai acheté un grand bateau à la foire nautique qui a eu lieu à Montréal plus tôt cette année et j’ai pu l’essayer cet été dans les eaux du lac Massawippi lors d’un récent voyage dans les Cantons de l’Est.
After: J’ai acheté un grand bateau à la foire nautique qui a eu lieu à Montréal plus tôt cette année. J’ai pu l’essayer cet été dans les eaux du lac Massawippi lors d’un récent voyage dans les Cantons de l’Est.
The task should be accomplished using deep learning and has to work both in french and in english.

Preventing Risk for Metabolic Syndrome in Workaholics: An Intervention

Tendencies towards workaholism have been linked to poor health and increased risk for diabetes and other chronic condition. A health improvement program that is interwoven within the workplace and leverages the ubiquitous use of smartphones has good potential of benefiting the workforce. The aim of this research project is to evaluate Transform, a digital health program created by Blue Mesa Health. The program is designed to prevent diabetes by helping people adopt healthier lifestyles. More specifically, this study will look at the impact Transform has on weight, physical activity workplace performance and stress management. Blue Mesa Health seeks to develop a strong research portfolio on Transform in order to provide a high-quality intervention that is both effective and competitive. BMH has plans to begin serving Canadians in 2019. Establishing collaborative partnerships in the Canadian academic arena is valuable to the growth and success of the company.

Applied next generation AI accelerator algorithm hardware co-optimization: using quantization, sparsity and hardware constraints during neural net training

This work aims to explore software and hardware co-optimization for deep neural network (DNN) inference applications. Once a model is trained to sufficient accuracy, the model is used to make inference or predictions based on this trained model. With increasing performance, more people are using these models for tasks such as translation, self-driving cars and speech recognition. This has greatly increased the demand for high performance inference hardware. The goal for this project is to investigate novel techniques to reduce latency and power consumption during inference while maintaining the same model accuracy.

Link predicting in court

The company Lexum is an undisputed leader in the development of information retrieval tools for the law – statutes, regulations and decisions of courts and tribunals. The project is to improve a new tool offer by the company. The tool is used to retrieve a list of legal subjects from a factual description. With that list extract, the tool provides a list of potential related document.