Machine Learning Operations (MLOps) Pipeline for Large-Scale Anomaly Detection in Derivative Markets

When we want to deploy the trained model in the production environment, we should make sure that the model is working properly. Thus,
we aim to extend the “data drift detection” and “model monitoring” components to work on large-scale datasets in production. Besides, we
aim to enable model versioning to make sure that the pipeline is flexible to replace the old models with the state-of-art models in the
production.

Faculty Supervisor:

Heng Li;Foutse Khomh;Maxime Lamothe

Student:

Partner:

Bourse de Montréal

Discipline:

Computer science

Sector:

Finance and Insurance

University:

Polytechnique Montréal

Program:

Accelerate

Current openings

Find the perfect opportunity to put your academic skills and knowledge into practice!

Find Projects