Summary: While Generative Adversarial Networks (GANs) have become the desired choice for generative tasks in the community, they also suffer from a nagging issue of mode collapse (cf. The current literature also has some empirical ways to handle this issue (cf. They present the technique of packing, in which the discriminator now uses multiple samples in its task. Detailed Comments: Clarity: The paper is very well written, both rigor and intuitive expositions are presented. Originality: As explained above in summary, perhaps this is the first time a framework of mode collapse is constructed and its theoretical underpinnings are discussed.

The model consists of a novel covariance function, which considers nodes in the 1-hop neighbourhood.

The paper presents a novel approach to alter the artistic style of images. This is achieved by combining an unsupervised style transfer method [11] with archetypal analysis [3] to learn style representations of collections of paintings (style images). Archetypes are computed for GanGogh and Vincent van Gogh paintings to learn style characteristics, which allows different stylization effects by changing the latent space of the archetypical representation. Due to the archetypical style representation, style changes remain interpretable. The style transfer is done in a hierarchical fashion similar to [11] by matching the first and second order statistics of the content and style feature maps (introduced as whitening and coloring transformations in [11]).

Neural Information Processing Systems http://nips.cc/

A Triangle Generative Adversarial Network ( -GAN) is developed for semisupervised cross-domain joint distribution matching, where the training data consists of samples from each domain, and supervision of domain correspondence is provided by only a few paired samples.

We propose a framework that learns a representation transferable across different domains and tasks in a label efficient manner.

We propose an efficient method to estimate the accuracy of classifiers using only unlabeled data. We consider a setting with multiple classification problems where the target classes may be tied together through logical constraints. For example, a set of classes may be mutually exclusive, meaning that a data instance can belong to at most one of them. The proposed method is based on the intuition that: (i) when classifiers agree, they are more likely to be correct, and (ii) when the classifiers make a prediction that violates the constraints, at least one classifier must be making an error. Experiments on four real-world data sets produce accuracy estimates within a few percent of the true accuracy, using solely unlabeled data. Our models also outperform existing state-of-the-art solutions in both estimating accuracies, and combining multiple classifier outputs. The results emphasize the utility of logical constraints in estimating accuracy, thus validating our intuition.

Neural Information Processing Systems http://nips.cc/

Generative adversarial networks (GANs) can implicitly learn rich distributions over images, audio, and data which are hard to model with an explicit likelihood. We present a practical Bayesian formulation for unsupervised and semi-supervised learning with GANs. Within this framework, we use stochastic gradient Hamiltonian Monte Carlo to marginalize the weights of the generator and discriminator networks. The resulting approach is straightforward and obtains good performance without any standard interventions such as label smoothing or mini-batch discrimination. By exploring an expressive posterior over the parameters of the generator, the Bayesian GAN avoids mode-collapse, produces interpretable and diverse candidate samples, and provides state-of-the-art quantitative results for semi-supervised learning on benchmarks including SVHN, CelebA, and CIFAR-10, outperforming DCGAN, Wasserstein GANs, and DCGAN ensembles.

Learning conditional distributions $\pi^*(\cdot|x)$ is a central problem in machine learning, which is typically approached via supervised methods with paired data $(x,y) \sim \pi^*$. However, acquiring paired data samples is often challenging, especially in problems such as domain translation. This necessitates the development of $\textit{semi-supervised}$ models that utilize both limited paired data and additional unpaired i.i.d. samples $x \sim \pi^*_x$ and $y \sim \pi^*_y$ from the marginal distributions. The usage of such combined data is complex and often relies on heuristic approaches. To tackle this issue, we propose a new learning paradigm that integrates both paired and unpaired data $\textbf{seamlessly}$ through the data likelihood maximization techniques. We demonstrate that our approach also connects intriguingly with inverse entropic optimal transport (OT). This finding allows us to apply recent advances in computational OT to establish a $\textbf{light}$ learning algorithm to get $\pi^*(\cdot|x)$. Furthermore, we demonstrate through empirical tests that our method effectively learns conditional distributions using paired and unpaired data simultaneously.