Latent Variable Modeling for Generative Concept Representations and Deep Generative Models
This work addresses the problem of improving interpretability and control in generative models for researchers and practitioners, but it appears incremental as it reviews and discusses existing concepts without introducing new methods.
The paper investigates latent variable modeling in deep generative models like VAEs and GANs, focusing on how latent spaces can be designed to support useful generative concept representations such as interpolation and attribute vectors, but it does not report specific results or numbers.
Latent representations are the essence of deep generative models and determine their usefulness and power. For latent representations to be useful as generative concept representations, their latent space must support latent space interpolation, attribute vectors and concept vectors, among other things. We investigate and discuss latent variable modeling, including latent variable models, latent representations and latent spaces, particularly hierarchical latent representations and latent space vectors and geometry. Our focus is on that used in variational autoencoders and generative adversarial networks.