Knowledge graphs store facts using relations between pairs of entities. In this work, we address the question of link prediction in knowledge graphs. Our general approach broadly follows neighborhood aggregation schemes such as that of Graph Convolutional Networks (GCN), which in turn was motivated by spectral graph convolutions. Our proposed model will aggregate information from […]

Automatically assessing a pilot performance during a flight training session is a capability that can enhance the flight instructor during his duty. From data gathered during a flight maneuver, we are looking for a way to automatically assess pilot performance to augment instructor performance and provide objectivity during flight training assessment.

This fundamental research project investigates semantic visual navigation tasks, such as asking a household robot to “go find my keys”. We seek to enhance the efficacy of repeated search tasks within the same environment, by explicitly building, maintaining, and exploiting a map of locations that the robot had previously explored. We also seek to exploit […]

API usability specifies how easy, efficient, error-preventing, and pleasant an API of a software library is from its users’ perspective. With machine learning (ML) techniques becoming increasingly powerful and pervasive, many non-programmers and casual users (e.g. domain experts in medicine or geography) started to explore ML libraries. However, many find them challenging to use because […]

In this project we aim to automatically learn an augmenter network by using an approximate online bilevel optimization procedure. We plan to learn a augmenter network that generates a distribution of transformations that minimizes the loss on a validation set. By unfolding the gradients of the training loss, we will optimize the loss on validation […]

The research project entails finding a method to detect corroded steel wires used in the manufacturing of tires at Michelin in Bridgewater. This would be achieved using methods that are proven safe for humans and the environment. Also, the detection equipment will be integrated in the existing Michelin facilities in Bridgwater, NS, Canada.

This project aims at investigating the effect of communication constraints on cooperative manipulators. Afterward, the communication constraints will be detailed modeled, based on that the control strategy specifically for the multi-robot system will be developed. The stability of the system under the designed control strategy should be guaranteed, and the performance regarding the execution of […]

The Mobile Robotics and Autonomous Systems Laboratory of Polytechnique Montréal is specialized in drone development and applications. The intern will join the team in charge of developing a payload transport system with drones. IN this research work we will develop a simulation environment, validate the different control strategies in the simulation environment and implement the […]

Machine learning (ML) has recently achieved impressive success in many applications. As ML starts to penetrate into safety-critical domains, security/robustness concerns on ML systems have received lots of attention lately. Very surprisingly, recent work has shown that current ML models are vulnerable to adversarial attacks, e.g. by perturbing the input slightly ML models can be […]

My proposed research project is devoted to Particle-in-cell (PIC) computer simulations to study the plasma waves instabilities in the exhaust plume of the Variable Specific Impulse Magnetoplasma Rocket (VASIMR) device. VASIMR is a type of electric propulsion plasma device, using the plasma as propellant, potentially for a long-term space mission with a much higher specific […]

Our project develops novel machine learning algorithms for interpreting complex, multi-agent scenarios in sports analytics. The collaboration with our industrial partner SPORTLOGiQ will tackle open problems in deep reinforcement learning to build novel capabilities in sports analytics for ice hockey. Deep reinforcement learning is a breakthrough technology with prominent successes in games such as Go […]

In this project, we propose a continual learning approach to face the problem of catastrophic forgetting in online image classification problems. Concretely, we propose a model that learns how to mask a series of general modules in a deep learning architecture, so that generalization emerges through the composition of those modules. This is of vital […]