Machine learning approaches for event prediction, relation modeling, and inference

Machine learning approaches are transforming fields such as finance, healthcare, electronic commerce, social networks, and natural disaster forecasting. We propose collaborative research that develops novel methods and applications of machine learning techniques for event prediction, modeling relations between entities, and inference techniques that can impact these domains. In the context of event prediction, we will develop methods based on the point process framework. We will develop novel models for learning the temporal distribution of human activities in streaming data (e.g., videos and person trajectories). Methods based on an integrated framework of neural networks and temporal point processes will be considered. For the problem of modeling relationships, we will build relational representations of entities, given graph structures describing potential interactions. Both supervised and unsupervised learning paradigms can be potentially utilized. Finally, we consider inference techniques for structured random variable spaces using deep learning approaches. TO BE CONT’D

Faculty Supervisor:

Fred Popowich

Student:

Thibaut Durand

Partner:

Borealis AI

Discipline:

Computer science

Sector:

Information and communications technologies

University:

Program: