Open AccessProceedings Article
Improved training of wasserstein GANs
Ishaan Gulrajani,Faruk Ahmed,Martin Arjovsky,Vincent Dumoulin,Aaron Courville +4 more
- Vol. 30, pp 5769-5779
TLDR
The authors proposed to penalize the norm of the gradient of the critic with respect to its input to improve the training stability of Wasserstein GANs and achieve stable training of a wide variety of GAN architectures with almost no hyperparameter tuning.Abstract:
Generative Adversarial Networks (GANs) are powerful generative models, but suffer from training instability. The recently proposed Wasserstein GAN (WGAN) makes progress toward stable training of GANs, but sometimes can still generate only poor samples or fail to converge. We find that these problems are often due to the use of weight clipping in WGAN to enforce a Lipschitz constraint on the critic, which can lead to undesired behavior. We propose an alternative to clipping weights: penalize the norm of gradient of the critic with respect to its input. Our proposed method performs better than standard WGAN and enables stable training of a wide variety of GAN architectures with almost no hyperparameter tuning, including 101-layer ResNets and language models with continuous generators. We also achieve high quality generations on CIFAR-10 and LSUN bedrooms.read more
Citations
More filters
Posted Content
Orthogonal Estimation of Wasserstein Distances
TL;DR: A new variant of sliced Wasserstein distance is proposed, the use of orthogonal coupling in Monte Carlo estimation of Wasserenstein distances is studied, and connections with stratified sampling are drawn.
Posted Content
Uncertainty Estimation Using a Single Deep Deterministic Neural Network
TL;DR: DUQ as discussed by the authors is a method for training a deterministic deep model that can find and reject out of distribution data points at test time with a single forward pass using RBF networks.
Journal ArticleDOI
Improved ArtGAN for Conditional Synthesis of Natural Image and Artwork
TL;DR: ArtGAN as discussed by the authors proposes a series of new approaches to improve generative adversarial network (GAN) for conditional image synthesis and name the proposed model as “ArtGAN. ” One of the key innovation of ArtGAN is that, the gradient of the loss function w.r.t. the label (randomly assigned to each generated image) is backpropagated from the categorical discriminator to the generator.
Proceedings ArticleDOI
SRDAN: Scale-aware and Range-aware Domain Adaptation Network for Cross-dataset 3D Object Detection
Weichen Zhang,Wen Li,Dong Xu +2 more
TL;DR: Wang et al. as discussed by the authors proposed a scale-aware and range-aware domain adaptation network (SRDAN), which takes advantage of the geometric characteristics of 3D data (i.e., size and distance) to guide the distribution alignment between two domains.
Posted Content
Generative Adversarial Zero-Shot Relational Learning for Knowledge Graphs
TL;DR: A novel formulation of zero-shot learning is considered, which is model-agnostic that could be potentially applied to any version of KG embeddings, and consistently yields performance improvements on NELL and Wiki dataset.
References
More filters
Dissertation
Learning Multiple Layers of Features from Tiny Images
TL;DR: In this paper, the authors describe how to train a multi-layer generative model of natural images, using a dataset of millions of tiny colour images, described in the next section.
Journal ArticleDOI
Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning
TL;DR: This article presents a general class of associative reinforcement learning algorithms for connectionist networks containing stochastic units that are shown to make weight adjustments in a direction that lies along the gradient of expected reinforcement in both immediate-reinforcement tasks and certain limited forms of delayed-reInforcement tasks, and they do this without explicitly computing gradient estimates.
Posted Content
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
TL;DR: This work introduces a class of CNNs called deep convolutional generative adversarial networks (DCGANs), that have certain architectural constraints, and demonstrates that they are a strong candidate for unsupervised learning.
Posted Content
Improved Techniques for Training GANs
TL;DR: In this article, the authors present a variety of new architectural features and training procedures that apply to the generative adversarial networks (GANs) framework and achieve state-of-the-art results in semi-supervised classification on MNIST, CIFAR-10 and SVHN.
Proceedings Article
Categorical Reparameterization with Gumbel-Softmax
Eric Jang,Shixiang Gu,Ben Poole +2 more
TL;DR: Gumbel-Softmax as mentioned in this paper replaces the non-differentiable samples from a categorical distribution with a differentiable sample from a novel Gumbel softmax distribution, which has the essential property that it can be smoothly annealed into the categorical distributions.