Dropout-GAN: Learning from a Dynamic Ensemble of Discriminators
By Gonçalo Mordido, Haojin Yang, Christoph Meinel. 2018
54 citations recorded by the source corpus.
We propose to incorporate adversarial dropout in generative multi-adversarial networks, by omitting or dropping out, the feedback of each discriminator in the framework with some probability at the end of each batch. Our approach forces the single generator not to constrain its output to satisfy a single discriminator, but, instead, to satisfy a dynamic ensemble of discriminators. We show that this leads to a more generalized generator, promoting variety in the generated samples and avoiding the common mode collapse problem commonly experienced with generative adversarial networks (GANs). We further provide evidence that the proposed framework, named Dropout-GAN, promotes sample diversity both within and across epochs, eliminating mode collapse and stabilizing training.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Generative Models
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Bernt Schiele
- Diane Joyce Cook
- Garrett Wilson
- Edgar Schönfeld
- Peng Wang
- Samuli Laine
- Timo Aila
- Jiatao Gu
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Bernt Schiele
- Ming-Hsuan Yang
- Dacheng Tao
- Jitendra Malik
- Trevor Darrell
- Xiaogang Wang
- Andrea Vedaldi
In Generative Models
- Tero Karras
- Timo Aila
- Samuli Laine
- Anna Khoreva
- Eli Shechtman
- Richard Zhang
- Jun-Yan Zhu
- Christian Theobalt
Citations in the corpus
- A Style-Based Generator Architecture for Generative Adversarial Networks (2018, methodology): “, using multiple discriminators [17, 43, 10], multiresolution discrimination [55, 51], or self-attention [57].”
- A U-Net Based Discriminator for Generative Adversarial Networks (2020, background): “Other lines of work focused mainly on improving the discriminator by exploiting multiple [36, 13, 11] and multi-resolution [45, 42] discriminators, using spatial feedback of the discriminator [17], an auto-encoder architecture with the reconstructionbased feedback to the generator [52] or self-super”
- StyleNeRF: A Style-based 3D-Aware Generator for High-resolution Image Synthesis (2021, background): “(2014), GANs have demonstrated high-quality results (Durugkar et al., 2017; Mordido et al., 2018; Doan et al., 2019; Zhang et al., 2019; Brock et al., 2018; Karras et al., 2018).”
- A Survey of Unsupervised Deep Domain Adaptation (2018, methodology): “Many methods have been proposed to resolve these training challenges using a variety of tricks [81, 90, 179, 207, 219, 229], network architecture choices [116, 190, 207], objective modifications [5, 15, 66, 87, 112, 120, 161, 163, 165, 176–178, 283], mixtures or ensembles [6, 63, 78, 93, 117, 170, 1”
- Generative Adversarial Networks (GANs) (2020, methodology): “Training of Multiple Discriminators D2GAN [66], GMAN [67], Stabilizing GAN [68], Dropout GAN [69], SGAN [70]”