REVIEW 15 cited by
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
The relativistic discriminator: a key element missing from standard GAN
read the original abstract
In standard generative adversarial network (SGAN), the discriminator estimates the probability that the input data is real. The generator is trained to increase the probability that fake data is real. We argue that it should also simultaneously decrease the probability that real data is real because 1) this would account for a priori knowledge that half of the data in the mini-batch is fake, 2) this would be observed with divergence minimization, and 3) in optimal settings, SGAN would be equivalent to integral probability metric (IPM) GANs. We show that this property can be induced by using a relativistic discriminator which estimate the probability that the given real data is more realistic than a randomly sampled fake data. We also present a variant in which the discriminator estimate the probability that the given real data is more realistic than fake data, on average. We generalize both approaches to non-standard GAN loss functions and we refer to them respectively as Relativistic GANs (RGANs) and Relativistic average GANs (RaGANs). We show that IPM-based GANs are a subset of RGANs which use the identity function. Empirically, we observe that 1) RGANs and RaGANs are significantly more stable and generate higher quality data samples than their non-relativistic counterparts, 2) Standard RaGAN with gradient penalty generate data of better quality than WGAN-GP while only requiring a single discriminator update per generator update (reducing the time taken for reaching the state-of-the-art by 400%), and 3) RaGANs are able to generate plausible high resolutions images (256x256) from a very small sample (N=2011), while GAN and LSGAN cannot; these images are of significantly better quality than the ones generated by WGAN-GP and SGAN with spectral normalization.
Forward citations
Cited by 15 Pith papers
-
Language-Assisted Super-Resolution from Real-World Low-Resolution Patches
LA-SR redefines unpaired super-resolution in language space by projecting images into a semantically rich representation and applying vision-language model guided losses to handle real-world degradations extracted fro...
-
Learning Stratigraphically Consistent Relative Geologic Time from 3D Seismic Data via Sinusoidal Mapping
RGT-Est transforms RGT estimation into a sinusoidal space with joint losses to capture fine horizons and global stratigraphic order from seismic data, outperforming prior AI methods especially with sparse horizon priors.
-
Learning Stratigraphically Consistent Relative Geologic Time from 3D Seismic Data via Sinusoidal Mapping
RGT-Est reformulates relative geologic time learning via sinusoidal mapping and joint pointwise-perceptual-adversarial losses to enforce stratigraphic consistency, outperforming prior AI methods on field data especial...
-
Whether, Which, and Whose: Solving the Triple Challenge of Deepfake Proactive Forensics in Multi-Face Scenarios
DAWF introduces isolated identity attribution spaces and selective regional supervision to unify detection, localization, and source tracing for multi-face deepfakes.
-
Visualizing High-Dimensional Graph Embeddings via Informed Multi-View Projections
High-dimensional graph embeddings are projected to 2D views optimized for aesthetic metrics using a differentiable edge-crossing surrogate, outperforming standard layouts in tests.
-
Learning Stratigraphically Consistent Relative Geologic Time from 3D Seismic Data via Sinusoidal Mapping
RGT-Est transforms relative geologic time estimation into a sinusoidal space and applies pointwise, perceptual, and adversarial losses to achieve better stratigraphic consistency and horizon correlation on seismic data.
-
Whether, Which, and Whose: Solving the Triple Challenge of Deepfake Proactive Forensics in Multi-Face Scenarios
DAWF embeds identity watermarks via a parallel multi-face architecture and uses selective loss to answer which face was forged and whose identity was used.
-
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.
-
Continuous Adversarial Flow Models
Continuous adversarial flow models replace MSE in flow matching with adversarial training via a discriminator, improving guidance-free FID on ImageNet from 8.26 to 3.63 for SiT and similar gains for JiT and text-to-im...
-
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.
-
Language-Assisted Super-Resolution from Real-World Low-Resolution Patches
LA-SR extracts real LR patches from depth-varying regions in single images and uses vision-language models with linguistic content and quality losses for unpaired super-resolution.
-
BioVid: Autoregressive Video Generation with Biological Behavior Semantic Comprehension
BioVid is a data-driven autoregressive model using 2D-encode/3D-decode tokenization and causal Transformer with EOS termination that reproduces real action duration distributions (W1 distance 1.24 frames) on NTU RGB+D...
-
MaCo-GAN: Manifold-Contrastive Adversarial Learning for Single Image Super-Resolution
MaCo-GAN introduces a manifold-contrastive GAN that replaces adversarial loss with a contrastive minimax game over synthesized fake samples to improve the perception-distortion trade-off in SISR.
-
Potential and challenges of generative adversarial networks for super-resolution in 4D Flow MRI
Wasserstein GAN super-resolution recovers near-wall velocities in 4D Flow MRI with vNRMSE 6.9% versus 9.6% for non-adversarial baseline, though training stability depends on loss function choice.
-
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.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.