Fig. 7
From: Sparse multi-output Gaussian processes for online medical time series prediction
The estimated population-level basis kernels and corresponding Bq matrices for patients with neoplasms. We show the kernels estimated (a) without a sparse prior (Q′=5) and (b) with a sparse prior (Q′=5). The sparsity of the Bq matrices are calculated as the percentage of nearly zero entries (i.e., values ≤10−3). The units for length-scale or period are (d) for days and (h) for hours