Sparse, Geometric Autoencoder Models of V1

Open in new window