Sampling Generative Networks
This provides tools for researchers working with generative models to better analyze and interpret latent spaces, though it appears incremental as it builds on existing sampling and visualization approaches.
The paper tackles the problem of sampling and visualizing latent spaces in generative models by introducing techniques like spherical linear interpolation for sharper samples and new visualization methods like J-Diagrams and MINE grids, with examples shown on Variational Autoencoders and Generative Adversarial Networks.
We introduce several techniques for sampling and visualizing the latent spaces of generative models. Replacing linear interpolation with spherical linear interpolation prevents diverging from a model's prior distribution and produces sharper samples. J-Diagrams and MINE grids are introduced as visualizations of manifolds created by analogies and nearest neighbors. We demonstrate two new techniques for deriving attribute vectors: bias-corrected vectors with data replication and synthetic vectors with data augmentation. Binary classification using attribute vectors is presented as a technique supporting quantitative analysis of the latent space. Most techniques are intended to be independent of model type and examples are shown on both Variational Autoencoders and Generative Adversarial Networks.