Article: work-in-progress.
Scalable Bayesian regression with latent sparse Gaussian Processes. Sparsity is not achieved using a small number of knots/sensors/inducing points but rather using sparse Directed Acyclic Graphs.
Data application: HELIX exposome data.
Poster award at ISBA World Meeting 2022, Montreal, Canada.
Slides from the 2021 Exposome Challenge: https://www.isglobal.org/documents/10179/9117539/1_Topic5_Michele_Peruzzi.pdf/30c5e893-f2da-4452-b2c4-a3e075425b77
