Boosting Interpretability and Reducing Resource Demands of Machine Learning Models using Tensor Networks
The rapid recent rise in the use of machine learning (ML) tools in business and everyday life has introduced both opportunities and new challenges. While powerful, large-scale ML models come with massive energy demands, as well as output that is hard to predict or interpret. This project will focus on the application of tensor networks (TNs), a rich family of models and methods originating in quantum physics and applied mathematics, to alleviate these issues within state-of-the-art ML models.
Voir la description complète du projetGuillaume Rabusseau
Zapata Canada
Computer science
Professional, scientific and technical services
McGill University; Université de Montréal
Accelerate