This task has received a considerable amount of attention, particularly for classical ways of modeling distributions like structured prediction.

Inferring the most likely configuration for a subset of variables of a joint distribution given the remaining ones -- which we refer to as co-generation -- is an important challenge that is computationally demanding for all but the simplest settings. This task has received a considerable amount of attention, particularly for classical ways of modeling distributions like structured prediction. In contrast, almost nothing is known about this task when considering recently proposed techniques for modeling high-dimensional distributions, particularly generative adversarial nets (GANs). Therefore, in this paper, we study the occurring challenges for co-generation with GANs. To address those challenges we develop an annealed importance sampling based Hamiltonian Monte Carlo co-generation algorithm. The presented approach significantly outperforms classical gradient based methods on a synthetic and on the CelebA and LSUN datasets.

It would be nice to have this as a baseline (this baselines is even easier to implement than the method proposed in this paper). Other interesting examples might include ill-posed problems like inverting a Radon transform for medical data or even doing multiple things at once with one model. It would also be nice, if Fig.4 was referenced much earlier in Sec. It would be nice if the authors could elaborate on this point. Are there simple experiments that show this? * "We show MSSIM and MSE between the ground truth and the final output" (l.

This is a purely empirical study that considers a problem of co-generation in the context of deep unsupervised generative models. Given a part of the example is observed, one is required to fill in the remaining (unobserved) part in a reasonable way. The problem is well motivated by applications such as image in-painting. The authors provide an extensive overview of the existing literature. The proposed solution is simple and uses an already trained GAN generator G: Z \to X to find latent vectors z resulting in outputs G(z) looking similar to the observed part of the image.

Inferring the most likely configuration for a subset of variables of a joint distribution given the remaining ones -- which we refer to as co-generation -- is an important challenge that is computationally demanding for all but the simplest settings. This task has received a considerable amount of attention, particularly for classical ways of modeling distributions like structured prediction. In contrast, almost nothing is known about this task when considering recently proposed techniques for modeling high-dimensional distributions, particularly generative adversarial nets (GANs). Therefore, in this paper, we study the occurring challenges for co-generation with GANs. To address those challenges we develop an annealed importance sampling based Hamiltonian Monte Carlo co-generation algorithm.

Inferring the most likely configuration for a subset of variables of a joint distribution given the remaining ones -- which we refer to as co-generation -- is an important challenge that is computationally demanding for all but the simplest settings. This task has received a considerable amount of attention, particularly for classical ways of modeling distributions like structured prediction. In contrast, almost nothing is known about this task when considering recently proposed techniques for modeling high-dimensional distributions, particularly generative adversarial nets (GANs). Therefore, in this paper, we study the occurring challenges for co-generation with GANs. To address those challenges we develop an annealed importance sampling based Hamiltonian Monte Carlo co-generation algorithm. The presented approach significantly outperforms classical gradient based methods on a synthetic and on the CelebA and LSUN datasets.