Geometric instability of out of distribution data across autoencoder architecture. (arXiv:2201.11902v1 [cs.LG])

Web: http://arxiv.org/abs/2201.11902

We study the map learned by a family of autoencoders trained on MNIST, and
evaluated on ten different data sets created by the random selection of pixel
values according to ten different distributions. Specifically, we study the
eigenvalues of the Jacobians defined by the weight matrices of the autoencoder
at each training and evaluation point. For high enough latent dimension, we
find that each autoencoder reconstructs all the evaluation data sets as similar
\emph{generalized characters}, but that this reconstructed \emph{generalized …

architecture arxiv autoencoder data distribution