AGAN is the first neural architecture search method for GANs that discovers architectures outperforming state-of-the-art on CIFAR-10 unsupervised image generation and competitive on supervised tasks.
hub
Courville
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
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 low-quality 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 over discrete data. We also achieve high quality generations on CIFAR-10 and LSUN bedrooms.
hub tools
citation-role summary
method 2
background 1
citation-polarity summary
representative citing papers
JanusPipe introduces SymFold and WaveK to enable efficient 3D-parallel training for conservative MLIPs, reporting 1.51x and 1.45x average throughput gains over 1F1B and Hanayo baselines on 32 GPUs.
Progressive growing stabilizes GAN training to produce high-resolution images of unprecedented quality and achieves a record unsupervised inception score of 8.80 on CIFAR10.
MMD GANs have unbiased critic gradients but biased generator gradients from sample-based learning, and the Kernel Inception Distance provides a practical new measure for GAN convergence and dynamic learning rate adaptation.
A generative adversarial network emulator upscales low-resolution N-body simulations with non-zero curvature to high resolution, recovering most large-scale power but with up to 10% small-scale suppression and altered halo profiles.
A tractable ensemble distributionally robust Bayesian optimization method achieves improved sublinear regret bounds under context uncertainty.
A CUDA-parallel voxelizer and hierarchical SGGX clustering representation enable faster, more accurate level-of-detail rendering of sparse microgeometry volumes.
A GAN inversion method coupled with property prediction enables inverse design of NiTi-based SMAs, with experimental validation yielding an alloy at 404°C transformation temperature and 9.9 J/cm³ work output.
Experiments on real industrial time series show that partial model sharing improves diffusion model performance in bandwidth-limited non-IID settings, while full sharing stabilizes GAN training but offers less robustness than VAE or DDPM alternatives.
A conditional Wasserstein GAN generates complete kinematic event summaries for IBD-CC, NC, and NuEElastic electron neutrino interactions that match GENIE distributions in 1D marginals and correlations.
Semi-supervised Bayesian GANs with log-signatures for uncertainty-aware credit card fraud detection show consistent improvements over benchmarks on the BankSim simulator under varying label proportions.
Wasserstein GAN generates synthetic fraud transactions that improve classifier performance on credit card data more stably than standard or conditional GAN variants.
citing papers explorer
-
AGAN: Towards Automated Design of Generative Adversarial Networks
AGAN is the first neural architecture search method for GANs that discovers architectures outperforming state-of-the-art on CIFAR-10 unsupervised image generation and competitive on supervised tasks.
-
JanusPipe: Efficient Pipeline Parallel Training for Machine Learning Interatomic Potentials
JanusPipe introduces SymFold and WaveK to enable efficient 3D-parallel training for conservative MLIPs, reporting 1.51x and 1.45x average throughput gains over 1F1B and Hanayo baselines on 32 GPUs.
-
Progressive Growing of GANs for Improved Quality, Stability, and Variation
Progressive growing stabilizes GAN training to produce high-resolution images of unprecedented quality and achieves a record unsupervised inception score of 8.80 on CIFAR10.
-
Demystifying MMD GANs
MMD GANs have unbiased critic gradients but biased generator gradients from sample-based learning, and the Kernel Inception Distance provides a practical new measure for GAN convergence and dynamic learning rate adaptation.
-
Separate Universe Super-Resolution Emulator
A generative adversarial network emulator upscales low-resolution N-body simulations with non-zero curvature to high resolution, recovering most large-scale power but with up to 10% small-scale suppression and altered halo profiles.
-
Ensemble Distributionally Robust Bayesian Optimisation
A tractable ensemble distributionally robust Bayesian optimization method achieves improved sublinear regret bounds under context uncertainty.
-
Fast Voxelization and Level of Detail for Microgeometry Rendering
A CUDA-parallel voxelizer and hierarchical SGGX clustering representation enable faster, more accurate level-of-detail rendering of sparse microgeometry volumes.
-
Generative Inversion for Property-Targeted Materials Design: Application to Shape Memory Alloys
A GAN inversion method coupled with property prediction enables inverse design of NiTi-based SMAs, with experimental validation yielding an alloy at 404°C transformation temperature and 9.9 J/cm³ work output.
-
On the Tradeoffs of On-Device Generative Models in Federated Predictive Maintenance Systems
Experiments on real industrial time series show that partial model sharing improves diffusion model performance in bandwidth-limited non-IID settings, while full sharing stabilizes GAN training but offers less robustness than VAE or DDPM alternatives.
-
Conditional Wasserstein GAN for Simulating Neutrino Event Summaries using Incident Energy of Electron Neutrinos
A conditional Wasserstein GAN generates complete kinematic event summaries for IBD-CC, NC, and NuEElastic electron neutrino interactions that match GENIE distributions in 1D marginals and correlations.
-
Semi-Supervised Bayesian GANs with Log-Signatures for Uncertainty-Aware Credit Card Fraud Detection
Semi-supervised Bayesian GANs with log-signatures for uncertainty-aware credit card fraud detection show consistent improvements over benchmarks on the BankSim simulator under varying label proportions.
-
Improving Detection of Credit Card Fraudulent Transactions using Generative Adversarial Networks
Wasserstein GAN generates synthetic fraud transactions that improve classifier performance on credit card data more stably than standard or conditional GAN variants.