Papers › DuelGAN: A Duel Between Two Discriminators Stabilizes the GAN Training

DuelGAN: A Duel Between Two Discriminators Stabilizes the GAN Training

19 Jan 2021arXiv:2101.07524archive 2025-07-28

Jiaheng Wei, Minghao Liu, Jiahao Luo, Andrew Zhu, James Davis, Yang Liu

In this paper, we introduce DuelGAN, a generative adversarial network (GAN) solution to improve the stability of the generated samples and to mitigate mode collapse. Built upon the Vanilla GAN's two-player game between the discriminator D₁ and the generator G, we introduce a peer discriminator D₂ to the min-max game. Similar to previous work using two discriminators, the first role of both D₁, D₂ is to distinguish between generated samples and real ones, while the generator tries to generate high-quality samples which are able to fool both discriminators. Different from existing methods, we introduce another game between D₁ and D₂ to discourage their agreement and therefore increase the level of diversity of the generated samples. This property alleviates the issue of early mode collapse by preventing D₁ and D₂ from converging too fast. We provide theoretical analysis for the equilibrium of the min-max game formed among G, D₁, D₂. We offer convergence behavior of DuelGAN as well as stability of the min-max game. It's worth mentioning that DuelGAN operates in the unsupervised setting, and the duel between D₁ and D₂ does not need any label supervision. Experiments results on a synthetic dataset and on real-world image datasets (MNIST, Fashion MNIST, CIFAR-10, STL-10, CelebA, VGG, and FFHQ) demonstrate that DuelGAN outperforms competitive baseline work in generating diverse and high-quality samples, while only introduces negligible computation cost.

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Image GenerationVocal Bursts Valence Prediction

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Generation CIFAR-10 PeerGAN FID 21.55 #56 of 78 Archive leaderboard report
Image Generation CelebA 64x64 PeerGAN FID 13.95 #30 of 39 Archive leaderboard report
Image Generation Fashion-MNIST PeerGAN FID 21.73 #3 of 7 Archive leaderboard report
Image Generation STL-10 PeerGAN FID 51.37 #29 of 31 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

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