In this project I will develop a computational framework for the automatic detection of task-related biomarkers in PD patients during speech production. This framework will help us to predict the timing of electrical stimulation in STN with the purpose of understanding the role of STN in encoding different aspects of speech production. To achieve this […]

Given a group of molecules and a specification of reaction conditions, do chemical reactions occur? If so, what products are produced, and what is their relative abundance? This problem pervades chemistry, with applications in environmental science (e.g., the degradation of pollutants), molecular sensing (e.g., interpreting the results of tandem mass spectrometry), chemical synthesis (e.g., finding […]

Avec le développement en parallèle de technologies de traitement automatique des langues de plus en plus performantes, de théories linguistiques de plus en plus précises et de banque de données de plus en plus riches, il est maintenant plus facile de comparer les langues et d’identifier leurs points communs et leurs divergences. Un défi pour […]

Metal additive manufacturing is a promising manufacturing technique that has attracted attentions in the recent years due to the ability to manufacture complex parts. During the process, metal goes through complex thermal treatment, which causes defects such as porosity and lack of fusion in the final part. Understanding the relationship between process parameters and parts’ […]

Since 2003, Humanware develops a GPS-based dedicated device that helps blind people to orient themselves and guides them to their destination. In this project, we want to enhance the user experience by providing a better understanding of the environment and interacting with the device in a more natural way. This project specifically addresses how the […]

Big data refers to a class of applications that operate on large amounts of data. One paradigm that fits the big data applications category is the Internet of Things (IoT). Millions of IoT devices continuously produce data and exchange information to support critical applications in different scenarios, such as smart cities and smart homes. As […]

Many manufacturing companies face production planning problems. Decisions that need to be taken over multiple periods include when and how much to produce of specific products, so that demand is satisfied. In this project we consider a complex production process, in which items with a customized length need to be cut from bigger objects with […]

Additive manufacturing (AM) is a product construction process by adding successive layers of materials according to computer-aided design models. Recently, metal AM products have been increasingly applied in aerospace, automotive, and biomedical industries. However, the complex interactions (i.e. thermal field) between metallic components and manufacturing processes make the product qualities inconsistent and unreliable, resulting in […]

Haptic signals are sent to movie or home theater motorized seats in order to create an immersive environment to the viewer by applying movements and vibration to the seat. The haptic signals are currently created using a tedious manual design process. These signals need to be matched and synchronized to specific actions in the movie, […]

Pipeline alerts are beneficial to analysts to determine where their attention is needed. However, a high false positive rate leads to a noisy stream and wastes analysts’ time. The first sub-project will aim to classify the alerts as either low or high likelihood of a false positive, allowing analysts to spend their time where it […]

The products manufactured at GE Bromont go through a series of processes that produce the specific characteristics of each product according to required specifications. The number of these characteristics is very large, and several characteristics are dependent on the others. This reality makes the use of traditional statistical techniques for quality control of products laborious […]

The main objective of this project is to investigate, develop and evaluate state-of-the-art deep-learning algorithms for joint few-shot classification and out-of-distribution (OOD) detection. Few-shot learning deals with the challenges of limited supervision, and OOD detection attempts to identify inputs that do not belong to the set of classes seen during training. The two research problems […]