Abstract
The neural architecture search (NAS) method provides a new approach to the design of generative adversarial network (GAN)’s architecture. The existing state-of-the-art algorithm, AutoGAN, is an example in discovering GAN’s architecture using reinforcement learning (RL)-based NAS. However, performance differences between the candidate architectures are not taken into consideration. In this paper, one new ImprovedAutoGAN algorithm based on AutoGAN is proposed. We found that which candidate architectures to use can affect the performance of the final network. So in the process of architecture search, compared with AutoGAN, in which candidate architectures are selected randomly, our method uses gradient bandit algorithm to increase the probability of selecting networks with better performance. This paper also introduces a temperature coefficient in the algorithm to prevent the search results from getting trapped in the local optimum. The GANs are searched using the same search space as AutoGAN, and the discovered GAN has a Frechet inception distance (FID) score of 11.60 on CIFAR-10, reaching the best level in the current RL-based NAS methods. Experiments also show that the transferability of this GAN is satisfying.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Notes
In Liu et al. (2018), the term “cell” is called “block.” Here, we use “cell” instead in order to be consistent with AutoGAN.
In stage 1, the generator is composed of one cell, so there is no candidate architectures and the operation to select it.
References
Abualigah LM, Khader AT, Hanandeh ES (2018a) Hybrid clustering analysis using improved krill herd algorithm. Appl Intell 48(11):4047–4071
Abualigah LMQ (2019) Feature selection and enhanced krill herd algorithm for text document clustering. Springer, Berlin
Abualigah LM, Khader AT, Hanandeh ES (2018b) A new feature selection method to improve the document clustering using particle swarm optimization algorithm. J Comput Sci 25:456–466
Brock A, Lim T, Ritchie JM, Weston N (2017) Smash: one-shot model architecture search through hypernetworks. In: International conference on learning representations (ICLR)
Brock A, Donahue J, Simonyan K (2019) Large scale gan training for high fidelity natural image synthesis. In: International conference on learning representations (ICLR)
Cai H, Zhu L, Han S (2019) Proxylessnas: direct neural architecture search on target task and hardware. In: International conference on learning representations (ICLR)
Elsken T, Metzen JH, Hutter F (2018) Efficient multi-objective neural architecture search via lamarckian evolution. In: International conference on learning representations (ICLR)
Gao C, Chen Y, Liu S, Tan Z, Yan S (2019) Adversarialnas: adversarial neural architecture search for gans. arXiv preprint arXiv:1912.02037
Gong X, Chang S, Jiang Y, Wang Z (2019) Autogan: neural architecture search for generative adversarial networks. In: IEEE International conference on computer vision (ICCV), pp 3224–3234
Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial nets. In: Advances in neural information processing systems (NIPS), pp 2672–2680
Grinblat GL, Uzal LC, Granitto PM (2017) Class-splitting generative adversarial networks. arXiv preprint arXiv:1709.07359
Gulrajani I, Ahmed F, Arjovsky M, Dumoulin V, Courville AC (2017) Improved training of wasserstein gans. In: Advances in neural information processing systems (NIPS), pp 5767–5777
He H, Wang H, Lee GH, Tian Y (2019) Probgan: towards probabilistic gan with theoretical guarantees. In: International conference on learning representations (ICLR)
Heusel M, Ramsauer H, Unterthiner T, Nessler B, Hochreiter S (2017) Gans trained by a two time-scale update rule converge to a local nash equilibrium. In: Neural information processing systems (NIPS), pp 6626–6637
Hoang Q, Nguyen TD, Le T, Phung D (2018) Mgan: training generative adversarial nets with multiple generators. In: International conference on learning representations (ICLR)
Karras T, Aila T, Laine S, Lehtinen J (2017) Progressive growing of gans for improved quality, stability, and variation. In: International conference on learning representations (ICLR)
Liu C, Zoph B, Neumann M, Shlens J, Hua W, Li LJ, Fei-Fei L, Yuille A, Huang J, Murphy K (2018) Progressive neural architecture search. In: Proceedings of the european conference on computer vision (ECCV), pp 19–34
Liu H, Simonyan K, Yang Y (2019) Darts: differentiable architecture search. In: International conference on learning representations (ICLR)
Luo R, Tian F, Qin T, Chen E, Liu TY (2018) Neural architecture optimization. In: Advances in neural information processing systems (NIPS), pp 7816–7827
Miyato T, Kataoka T, Koyama M, Yoshida Y (2018) Spectral normalization for generative adversarial networks. In: International conference on learning representations (ICLR)
Negrinho R, Gordon G (2017) Deeparchitect: automatically designing and training deep architectures. arXiv preprint arXiv:1704.08792
Perez-Rua JM, Baccouche M, Pateux S (2018) Efficient progressive neural architecture search. In: British machine vision conference (BMVC), p 150
Pham H, Guan MY, Zoph B, Le QV, Dean J (2018) Efficient neural architecture search via parameter sharing. In: International conference on machine learning (ICML)
Radford A, Metz L, Chintala S (2016) Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434
Real E, Moore S, Selle A, Saxena S, Suematsu YL, Tan J, Le QV, Kurakin A (2017) Large-scale evolution of image classifiers. In: International conference on machine learning (ICML), JMLR. org, vol 70, pp 2902–2911
Real E, Aggarwal A, Huang Y, Le QV (2019) Regularized evolution for image classifier architecture search. AAAI Conf Artif Intell 33:4780–4789
Salimans T, Goodfellow I, Zaremba W, Cheung V, Radford A, Chen X (2016) Improved techniques for training gans. In: Advances in neural information processing systems (NIPS), pp 2234–2242
So DR, Liang C, Le QV (2019) The evolved transformer. In: International conference on learning representations (ICLR), pp 5877–5886
Sutton RS, Barto AG (2011) Reinforcement learning: an introduction
Tran NT, Bui TA, Cheung NM (2018) Dist-gan: an improved gan using distance constraints. In: European conference on computer vision (ECCV), pp 370–385
Wang H, Huan J (2019) Agan: towards automated design of generative adversarial networks. arXiv preprint arXiv:1906.11080
Wang W, Sun Y, Halgamuge S (2018) Improving mmd-gan training with repulsive loss function. In: International conference on learning representations (ICLR)
Warde-Farley D, Bengio Y (2017) Improving generative adversarial networks with denoising feature matching. In: International conference on learning representations (ICLR)
Xie S, Zheng H, Liu C, Lin L (2019) Snas: stochastic neural architecture search. In: International conference on learning representations (ICLR)
Xu T, Zhang P, Huang Q, Zhang H, Gan Z, Huang X, He X (2018) Attngan: Fine-grained text to image generation with attentional generative adversarial networks. In: IEEE conference on computer vision and pattern recognition (CVPR), pp 1316–1324
Yang J, Kannan A, Batra D, Parikh D (2017) Lr-gan: layered recursive generative adversarial networks for image generation. In: International conference on learning representations (ICLR)
Zhang X, Huang Z, Wang N (2018) You only search once: single shot neural architecture search via direct sparse optimization. arXiv preprint arXiv:1811.01567
Zhang X, Wang Z, Liu D, Ling Q (2019) Dada: deep adversarial data augmentation for extremely low data regime classification. IEEE international conference on acoustics. Speech and Signal Processing (ICASSP), IEEE, pp 2807–2811
Zhong Z, Yan J, Wu W, Shao J, Liu CL (2018) Practical block-wise neural network architecture generation. In: IEEE conference on computer vision and pattern recognition (ICPR), pp 2423–2432
Zoph B, Le QV (2017) Neural architecture search with reinforcement learning. In: International conference on learning representations (ICLR)
Zoph B, Vasudevan V, Shlens J, Le QV (2018) Learning transferable architectures for scalable image recognition. In: IEEE conference on computer vision and pattern recognition (CVPR), pp 8697–8710
Acknowledgements
The authors are grateful to the College of Computer Science, Nanjing University of Posts and Telecommunications for kindly providing the required computational resources. We also thank Baofeng Zhang for his help with this Project. This study was funded by National Natural Science Foundation of China (NSFC) 2019-2022 (Grants 61877051 and 61872079). Associate Professor Jun Shen was also supported by research exchange program funded University Global Partnership Network.
Funding
This study was funded by National Natural Science Foundation of China (NSFC) 2019-2022 (Grants 61877051 and 61872079).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
This article does not contain any studies with human participants or animals performed by any of the authors.
Additional information
Communicated by V. Loia.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Fan, Y., Zhou, G., Shen, J. et al. Toward gradient bandit-based selection of candidate architectures in AutoGAN. Soft Comput 25, 4367–4378 (2021). https://doi.org/10.1007/s00500-020-05446-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-020-05446-x