Spectral Normalization for Generative Adversarial Networks
read the original abstract
One of the challenges in the study of generative adversarial networks is the instability of its training. In this paper, we propose a novel weight normalization technique called spectral normalization to stabilize the training of the discriminator. Our new normalization technique is computationally light and easy to incorporate into existing implementations. We tested the efficacy of spectral normalization on CIFAR10, STL-10, and ILSVRC2012 dataset, and we experimentally confirmed that spectrally normalized GANs (SN-GANs) is capable of generating images of better or equal quality relative to the previous training stabilization techniques.
This paper has not been read by Pith yet.
Forward citations
Cited by 42 Pith papers
-
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.
-
Approximation of Maximally Monotone Operators : A Graph Convergence Perspective
Any maximally monotone operator can be approximated in local graph convergence by continuous encoder-decoder networks, with structure-preserving versions that retain maximal monotonicity via resolvent parameterizations.
-
CAWI: Copula-Aligned Weight Initialization for Randomized Neural Networks
CAWI replaces standard random initialization of input-to-hidden weights in randomized neural networks with samples drawn from a data-fitted copula that preserves observed feature dependencies, yielding consistent accu...
-
Assessment of cloud and associated radiation fields from a GAN stochastic cloud subcolumn generator
A CVAE-GAN U-Net stochastic generator trained on CloudSat-CALIPSO data halves RMSE in cloud histograms and cuts global shortwave cloud radiative effect bias by a factor of three versus the Räisänen method.
-
Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data
Asymmetric Langevin Unlearning uses public data to suppress unlearning noise costs by O(1/n_pub²), enabling practical mass unlearning with preserved utility under distribution mismatch.
-
Curated Synthetic Data Doesn't Have to Collapse: A Theoretical Study of Generative Retraining with Pluralistic Preferences
Recursive generative retraining with pluralistic preferences converges to a stable diverse distribution that satisfies a weighted Nash bargaining solution.
-
Tessellations of Semi-Discrete Flow Matching
Semi-discrete Flow Matching produces terminal assignment regions that are topologically simple (open, simply connected, homeomorphic to the ball under assumption) yet geometrically distinct from optimal transport Lagu...
-
KANs need curvature: penalties for compositional smoothness
A curvature penalty for KANs, derived to respect compositional effects and equipped with a proven upper bound on full-model curvature, produces smoother activations while preserving accuracy.
-
Physics-informed, Generative Adversarial Design of Funicular Shells
A modified DCGAN with an auxiliary discriminator using the membrane factor generates stable, previously unseen funicular shells optimized for pure compression in three dimensions.
-
Neural Stochastic Differential Equations on Compact State Spaces: Theory, Methods, and Application to Suicide Risk Modeling
The authors derive drift and diffusion constraints plus a parameterization that forces neural SDE solutions to remain inside compact polyhedral domains, yielding better forecasts on real EMA suicide-risk datasets than...
-
Training Deep Learning Models with Norm-Constrained LMOs
Scion is a new stochastic LMO-based optimizer family that unifies existing methods, supports unconstrained problems, and delivers hyperparameter transferability plus speedups on nanoGPT training.
-
Equivariant score-based generative models provably learn distributions with symmetries efficiently
Equivariant SGMs achieve improved Wasserstein-1 generalization bounds on group-invariant distributions and learn the symmetrized score via equivariant vector fields without augmentation, with non-equivariant models in...
-
A Plug-and-Play Method for Guided Multi-contrast MRI Reconstruction based on Content/Style Modeling
PnP-CoSMo is a modular plug-and-play iterative reconstruction technique that disentangles content and style in multi-contrast MR images to guide reconstruction from reference scans without k-space training data.
-
Diffusion Models Beat GANs on Image Synthesis
Diffusion models with architecture improvements and classifier guidance achieve superior FID scores to GANs on unconditional and conditional ImageNet image synthesis.
-
Assessment of cloud and associated radiation fields from a GAN stochastic cloud subcolumn generator
A CVAE-GAN model trained on satellite cloud data generates stochastic subcolumns that reproduce observed bimodal overlap distributions and cut shortwave cloud radiative effect bias by a factor of three compared to the...
-
AdamO: A Collapse-Suppressed Optimizer for Offline RL
AdamO modifies Adam with an orthogonality correction to ensure the spectral radius of the TD update operator stays below one, providing a theoretical stability guarantee for offline RL.
-
Pairing Regularization for Mitigating Many-to-One Collapse in GANs
Pairing regularization mitigates intra-mode collapse in GANs by penalizing redundant latent-to-sample mappings, improving recall under collapse-prone conditions or precision under stabilized training.
-
Sim2Real-AD: A Modular Sim-to-Real Framework for Deploying VLM-Guided Reinforcement Learning in Real-World Autonomous Driving
Sim2Real-AD enables zero-shot transfer of CARLA-trained VLM-guided RL policies to full-scale vehicles, reporting 75-90% success rates in car-following, obstacle avoidance, and stop-sign scenarios without real-world RL...
-
Thermodynamically consistent machine learning model for excess Gibbs energy
HANNA is a thermodynamically consistent ML model for predicting excess Gibbs energy from molecular structures, trained on various binary mixture data and extended to multi-component mixtures using geometric projection.
-
Fast gradient-free activation maximization for neurons in spiking neural networks
A Tensor Train decomposition-based method enables efficient gradient-free activation maximization for neurons in spiking neural networks by searching generative model latent spaces.
-
Knowledge Distillation in Iterative Generative Models for Improved Sampling Speed
Denoising Student distills the multi-step denoising process of score-based and diffusion models into a single forward pass, matching GAN sampling speed while producing comparable sample quality on CIFAR-10, CelebA, an...
-
Perceptually Motivated Method for Image Inpainting Comparison
Subjective comparison of nine inpainting algorithms produces proposed objective metrics with high correlation to human perception of realism.
-
Dual Adversarial Learning with Attention Mechanism for Fine-grained Medical Image Synthesis
Dual-discriminator GAN with adversarial attention improves fine-grained medical image synthesis, especially in hard-to-generate tumor regions, and outperforms prior methods on brain tumor and CT-to-MRI tasks.
-
RED: A ReRAM-based Deconvolution Accelerator
RED introduces pixel-wise mapping and zero-skipping dataflow for ReRAM deconvolution acceleration, reporting 1.15x-3.69x speedup and 8%-88.36% energy reduction versus prior ReRAM accelerators.
-
DASGAN -- Joint Domain Adaptation and Segmentation for the Analysis of Epithelial Regions in Histopathology PD-L1 Images
DASGAN trains a segmentation network on semi-automatically labeled CK images via unpaired translation to PD-L1, enabling epithelium segmentation and TC score estimation without serial sections.
-
Deep Exemplar-based Video Colorization
A recurrent end-to-end network for exemplar-based video colorization that unifies semantic correspondence and color propagation with a temporal consistency loss.
-
Margin-Adaptive Confidence Ranking for Reliable LLM Judgement
Introduces a margin-adaptive confidence ranking method that learns an estimator from simulated diversity and derives margin-dependent generalization bounds for use in fixed-sequence testing of LLM-human agreement.
-
Pion: A Spectrum-Preserving Optimizer via Orthogonal Equivalence Transformation
Pion is an optimizer that preserves the singular values of weight matrices in LLM training by applying orthogonal equivalence transformations.
-
Constrained Stochastic Spectral Preconditioning Converges for Nonconvex Objectives
Proximal stochastic spectral preconditioning converges for nonconvex constrained objectives under heavy-tailed noise, with a variance-reduced version achieving faster rates and a refined analysis of Muon iterations.
-
Selective Prediction from Agreement: A Lipschitz-Consistent Version Space Approach
Selective prediction abstains unless all Lipschitz-consistent heads in the version space agree on a certified label for each pool point.
-
Visual Generation in the New Era: An Evolution from Atomic Mapping to Agentic World Modeling
Visual generation models are evolving from passive renderers to interactive agentic world modelers, but current systems lack spatial reasoning, temporal consistency, and causal understanding, with evaluations overemph...
-
Learning to Emulate Chaos: Adversarial Optimal Transport Regularization
Adversarial optimal transport objectives train neural emulators with improved long-term statistical fidelity on chaotic systems.
-
NTIRE 2026 Challenge on Short-form UGC Video Restoration in the Wild with Generative Models: Datasets, Methods and Results
The NTIRE 2026 challenge releases the KwaiVIR benchmark for short-form UGC video restoration and reports strong results from 12 teams using generative models on both subjective and objective tracks.
-
HTMuon: Improving Muon via Heavy-Tailed Spectral Correction
HTMuon modifies Muon to produce heavier-tailed updates and weight spectra via HT-SR theory, yielding up to 0.98 lower perplexity on LLaMA pretraining and serving as a plug-in for other Muon variants.
-
Unsupervised Detection of Spatiotemporal Anomalies in PMU Data Using Transformer-Based BiGAN
T-BiGAN integrates window-attention Transformers in a BiGAN to achieve ROC-AUC 0.95 and average precision 0.996 for unsupervised spatiotemporal anomaly detection in PMU data.
-
Privacy-Preserving Semantic Segmentation from Ultra-Low-Resolution RGB Inputs
A fully joint-learning framework enables semantic segmentation from ultra-low-resolution RGB inputs while achieving a favorable privacy-performance trade-off and successful robotic deployment.
-
Layer-wise Derivative Controlled Networks
ChainzRule with DREG regularization claims 15.5x fewer parameters than standard models, 23.1% lower peak gradient volatility on MNIST, and 70.17% accuracy on Yelp Full ordinal regression.
-
A Wasserstein GAN-based climate scenario generator for risk management and insurance: the case of soil subsidence
A conditional Wasserstein GAN generates plausible future SWI drought trajectories for French insurance risk management under climate change.
-
Diving Deeper into Underwater Image Enhancement: A Survey
A comprehensive survey of deep learning-based underwater image enhancement with systematic experimental comparison of algorithms on multiple datasets.
-
Neural Embedding for Physical Manipulations
Generative model with normalized pairwise distance constraint discovers output space topologies from sparse data and outperforms GANs and VAEs by avoiding mode collapse.
-
Incremental Concept Learning via Online Generative Memory Recall
Pseudo-rehearsal method with cGAN-generated old-concept samples, balanced online recall, and concept contrastive loss for class-incremental learning on MNIST, Fashion-MNIST and SVHN.
-
Advancing Intelligent Sequence Modeling: Evolution, Trade-offs, and Applications of State- Space Architectures from S4 to Mamba
A survey tracing the evolution of state-space models like S4 and Mamba, their efficiency trade-offs, and applications in NLP, vision, and other domains.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.