Methods › General › Regularization › GAN Feature Matching
GAN Feature Matching
Introduced by Tim Salimans et al. in Improved Techniques for Training GANs
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
Feature Matching is a regularizing objective for a generator in generative adversarial networks that prevents it from overtraining on the current discriminator. Instead of directly maximizing the output of the discriminator, the new objective requires the generator to generate data that matches the statistics of the real data, where we use the discriminator only to specify the statistics that we think are worth matching. Specifically, we train the generator to match the expected value of the features on an intermediate layer of the discriminator. This is a natural choice of statistics for the generator to match, since by training the discriminator we ask it to find those features that are most discriminative of real data versus data generated by the current model.
Letting 𝐟(𝐱) denote activations on an intermediate layer of the discriminator, our new objective for the generator is defined as: ||𝔼_(x∼p_(data)) 𝐟(𝐱) − 𝔼_(𝐳∼p_𝐳(𝐳))𝐟(G(𝐳))||²₂. The discriminator, and hence 𝐟(𝐱), are trained as with vanilla GANs. As with regular GAN training, the objective has a fixed point where G exactly matches the distribution of training data.
Papers archive 2025-07-28
16 shown of 16, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.
-
Stock market forecasting using DRAGAN and feature matching 6 Jan 2023 · 1 repository · arXiv:2301.05693
-
VocBench: A Neural Vocoder Benchmark for Speech Synthesis 6 Dec 2021 · 1 repository · arXiv:2112.03099
-
Feature Quantization Improves GAN Training 5 Apr 2020 · 2 repositories · arXiv:2004.02088Syntology ran 1 of 4 samples · 3 unverified · 1 pointer-only (licence)
-
Imbalanced Data Learning by Minority Class Augmentation using Capsule Adversarial Networks 5 Apr 2020 · 0 repositories · arXiv:2004.02182
-
Learning the Loss Functions in a Discriminative Space for Video Restoration 20 Mar 2020 · 0 repositories · arXiv:2003.09124
-
Image Fine-grained Inpainting 7 Feb 2020 · 3 repositories · arXiv:2002.02609
-
Semi-Supervised Self-Growing Generative Adversarial Networks for Image Recognition 11 Aug 2019 · 0 repositories · arXiv:1908.03850
-
Few-shot 3D Multi-modal Medical Image Segmentation using Generative Adversarial Learning 29 Oct 2018 · 1 repository · arXiv:1810.12241Syntology ran 0 of 6 samples · 6 unverified
-
A Generative Model of Textures Using Hierarchical Probabilistic Principal Component Analysis 16 Oct 2018 · 0 repositories · arXiv:1810.06892
-
Out-of-domain Detection based on Generative Adversarial Network 1 Oct 2018 · 0 repositories
-
Adversarial Training for Adverse Conditions: Robust Metric Localisation using Appearance Transfer 9 Mar 2018 · 0 repositories · arXiv:1803.03341
-
Novelty Detection with GAN 28 Feb 2018 · 0 repositories · arXiv:1802.10560
-
Bayesian GAN 1 Dec 2017 · 1 repository
-
Bayesian GAN 26 May 2017 · 4 repositories · arXiv:1705.09558
-
CVAE-GAN: Fine-Grained Image Generation through Asymmetric Training 29 Mar 2017 · 3 repositories · arXiv:1703.10155
-
Improved Techniques for Training GANs 10 Jun 2016 · 46 repositories · arXiv:1606.03498Syntology ran 1 of 2 samples · 1 unverified
Tasks archive 2025-07-28
20 shown of 35 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections