Unsupervised Anomaly Detection in multivariate Time Series Data

The enormous amount of data generated can be exploited using state-of-the-art AI algorithms to drive business decisions. However, a significant drawback of existing approaches is that the algorithms require a considerable amount of human effort and energy to prepare and annotate the data. Recent advances in deep learning and AI propose to solve this bottleneck using a paradigm referred to as 'Unsupervised' algorithms. In this project, we aim to draw inspiration from some breakthrough works in Unsupervised methods for timeseries data to help Rithmik Solutions reduce manual effort and quicken the process of ideation, model development and delivering actionable insights to customers.

Intern: 
Makesh Narsimhan Sreedhar
Superviseur universitaire: 
Yoshua Bengio
Province: 
Quebec
Partenaire: 
Secteur: 
Partner University: 
Discipline: