Fig. 1.

Feature selection. We train a deep learning model that takes all the views, estimates a shared low-dimensional representation $\mathbf{Z}$ that drives the variation across the views, and obtains nonlinear reconstructions ($G_1(\mathbf{Z})$,...,$G_D(\mathbf{Z})$) of the original views. We impose sparsity constraints on the reconstructions allowing us to identify a subset of variables for each view (${\mathbf{I}}_1$, ...,${\mathbf{I}}_D$) that approximate the original data