Fig 2. Hybrid model with shared spatial filters.

a,b. Schemata of SI model (a) and EC model (b) from [39]. The SI model branch consists of spatial and temporal convolutional layers, a fully connected (FC) layer and a nonlinear layer (see Methods). The EC model branch is a convolutional autoencoder, consisting of an encoder and a decoder network. In the hybrid model, the two branches were trained in parallel with shared spatial filters (all spatial filters were shared; red). Input_SI: 8-frame UV-green noise (t₁ … t₈); Output_SI: predicted GCL cell Ca²⁺ responses; Input_EC: UV-green natural images; Output_EC: reconstructed Input_EC. c. Example for the different inputs (natural images, phase-scrambled natural images, and noise) for the EC branch in hybrid models (hybrid-natural, hybrid-pha-scr, hybrid-noise). d. Using PCA filters as basis vectors for spatial convolutional filters of the SI model; SI-PCA learned 16 weight vectors ($\overrightarrow{w_1}\dots \overrightarrow{w_{16}}$) with same vector length as the number of PCA basis elements.