Spectral Normalization for Generative Adversarial Networks
Takeru Miyato, Toshiki Kataoka, Masanori Koyama et al. · arXiv (Cornell University) · 2018
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.
Cited by 3,171 works. doi:10.48550/arxiv.1802.05957 Open-access PDF
Authors
References
- Deep Residual Learning for Image Recognition
- ImageNet Large Scale Visual Recognition Challenge
- AutoResearch Lite: a reproducible paper-inspired experiment runner (Adam mechanism reproduction and convergence-speed comparison)
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- 80 Million Tiny Images: A Large Data Set for Nonparametric Object and Scene Recognition
- Conditional Generative Adversarial Nets
- An analysis of single-layer networks in unsupervised feature learning
- Improved Techniques for Training GANs
- Conditional Image Synthesis With Auxiliary Classifier GANs
- Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks