Paired Wasserstein Autoencoders for Conditional Sampling
Wasserstein distances greatly influenced and coined various types of generative neural network models. Wasserstein autoencoders are particularly notable for their mathematical simplicity and straight-forward implementation. However, their adaptation to the conditional case displays theoretical difficulties. As a remedy, we propose the use of two paired autoencoders. Under the assumption of an optimal autoencoder pair, we leverage the pairwise independence condition of our prescribed Gaussian latent distribution to overcome this theoretical hurdle. We conduct several experiments to showcase the practical applicability of the resulting paired Wasserstein autoencoders. Here, we consider imaging tasks and enable conditional sampling for denoising, inpainting, and unsupervised image translation. Moreover, we connect our image translation model to the Monge map behind Wasserstein-2 distances.
Code (0)
등록된 구현이 없습니다.
Tasks
DenoisingTranslationSimilar Papers 제목 키워드 기반
Conditional Sampling via Wasserstein Autoencoders and Triangular Transport
We present Conditional Wasserstein Autoencoders (CWAEs), a framework for conditional simulation that exploits low-dimensional structure in both the conditioned and the conditioning variables. The key idea is to modify a …
Y-Diagonal Couplings: Approximating Posteriors with Conditional Wasserstein Distances
In inverse problems, many conditional generative models approximate the posterior measure by minimizing a distance between the joint measure and its learned approximation. While this approach also controls the distance b…
Generative Class-conditional Autoencoders
Recent work by Bengio et al. (2013) proposes a sampling procedure for denoising autoencoders which involves learning the transition operator of a Markov chain. The transition operator is typically unimodal, which limits …
DenoisingSemi-Supervised Generative Learning via Latent Space Distribution Matching
We introduce Latent Space Distribution Matching (LSDM), a novel framework for semi-supervised generative modeling of conditional distributions. LSDM operates in two stages: (i) learning a low-dimensional latent space fro…
Image Super-ResolutionConditional Sampling of Variational Autoencoders via Iterated Approximate Ancestral Sampling
Conditional sampling of variational autoencoders (VAEs) is needed in various applications, such as missing data imputation, but is computationally intractable. A principled choice for asymptotically exact conditional sam…
Imputation