Abstract
Generative adversarial networks (GANs) have shown extraordinary performance in generating high quality samples in domains including image, video, and text. GANs therefore have great potential in learning complex probability distributions in high dimensional spaces. However, current methods often miss capturing some of the modes in the examples, known as the mode collapse problem. The reason for this issue can be traced to that the initial generated manifold fails to cover the whole data manifold, while the training process is hard to recover from this failure. In this paper, we propose GANs with supervision signal (SSGAN), which introduces a supervision signal to alleviate this issue. The supervision signal tells the generator an approximate output corresponding to the input noise, which ensures the generated manifold to be close to the data manifold. Therefore, the generator could be able to better capture the whole data distribution. We have conducted experiments on MNIST, CIFAR 10 and CelebA datasets. The results show that our method outperforms several SoTA approaches measured by the inception score, mode score, and the newly proposed matching score.
K. Zhang—Independent Researcher.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Arjovsky, M., Chintala, S., Bottou, L.: Wasserstein GAN. In: International Conference on Machine Learning (ICML) (2017)
Arora, S., Ge, R., Liang, Y., Ma, T., Zhang, Y.: Generalization and equilibrium in generative adversarial nets. In: International Conference on Machine Learning (ICML) (2017)
Che, T., Li, Y., Jacob, A.P., Bengio, Y., Li, W.: Mode regularized generative adversarial networks. In: International Conference on Learning Representation (ICLR) (2017)
Chen, X., Duan, Y., Houthooft, R., Schulman, J., Sutskever, I., Abbeel, P.: InfoGAN: interpretable representation learning by information maximizing generative adversarial nets. In: Advances in Neural Information Processing Systems (NIPS) (2016)
Denton, E.L., Chintala, S., Fergus, R., et al.: Deep generative image models using a Laplacian pyramid of adversarial networks. In: Advances in Neural Information Processing Systems (NIPS), pp. 1486–1494 (2015)
Goodfellow, I.J., et al.: Generative adversarial nets. In: Advances in Neural Information Processing Systems (NIPS), pp. 2672–2680 (2014)
Gulrajani, I., Ahmed, F., Arjovsky, M., Dumoulin, V., Courville, A.: Improved training of wasserstein GANs. In: Advances in Neural Information Processing Systems (NIPS) (2017)
Ho, J., Ermon, S.: Generative adversarial imitation learning. In: Advances in Neural Information Processing Systems (NIPS), pp. 4565–4573 (2016)
Krizhevsky, A., Hinton, G.: Learning multiple layers of features from tiny images (2009)
LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998)
Liu, H., Lafferty, J.D., Wasserman, L.A.: Sparse nonparametric density estimation in high dimensions using the rodeo. In: Proceedings of the 11th International Conference on Artificial Intelligence and Statistics, (AISTATS), pp. 283–290 (2007)
Liu, Z., Luo, P., Wang, X., Tang, X.: Deep learning face attributes in the wild. In: Proceedings of International Conference on Computer Vision (ICCV) (2015)
Mathieu, M., Couprie, C., LeCun, Y.: Deep multi-scale video prediction beyond mean square error. International Conference on Learning Representation (ICLR) (2016)
Metz, L., Poole, B., Pfau, D., Sohl-Dickstein, J.: Unrolled generative adversarial networks. In: International Conference on Learning Representation (ICLR) (2017)
Miyato, T., Kataoka, T., Koyama, M., Yoshida, Y.: Spectral normalization for generative adversarial networks. In: International Conference on Learning Representation (ICLR) (2018)
Radford, A., Metz, L., Chintala, S.: Unsupervised representation learning with deep convolutional generative adversarial networks. In: International Conference on Learning Representation (ICLR) (2016)
Salimans, T., Goodfellow, I., Zaremba, W., Cheung, V., Radford, A., Chen, X.: Improved techniques for training GANs. In: Advances in Neural Information Processing Systems (NIPS), pp. 2234–2242 (2016)
Srivastava, A., Valkov, L., Russell, C., Gutmann, M.U.: VEEGAN: reducing mode collapse in GANs using implicit variational learning. In: Advances in Neural Information Processing Systems (NIPS), pp. 3310–3320 (2017)
Tolstikhin, I., Gelly, S., Bousquet, O., Simon-Gabriel, C.J., Schölkopf, B.: AdaGAN: boosting generative models. In: Advances in Neural Information Processing Systems (NIPS) (2017)
Wu, J., Zhang, C., Xue, T., Freeman, B., Tenenbaum, J.: Learning a probabilistic latent space of object shapes via 3D generative-adversarial modeling. In: Advances in Neural Information Processing Systems (NIPS), pp. 82–90 (2016)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Zhang, K. (2021). On Mode Collapse in Generative Adversarial Networks. In: Farkaš, I., Masulli, P., Otte, S., Wermter, S. (eds) Artificial Neural Networks and Machine Learning – ICANN 2021. ICANN 2021. Lecture Notes in Computer Science(), vol 12892. Springer, Cham. https://doi.org/10.1007/978-3-030-86340-1_45
Download citation
DOI: https://doi.org/10.1007/978-3-030-86340-1_45
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-86339-5
Online ISBN: 978-3-030-86340-1
eBook Packages: Computer ScienceComputer Science (R0)