Ensembles of Generative Adversarial Networks

Ensembles are a popular way to improve results of discriminative CNNs. The combination of several networks trained starting from different initializations improves results significantly. In this paper we investigate the usage of ensembles of GANs. The specific nature of GANs opens up several new ways to construct ensembles. The first one is based on the fact that in the minimax game which is played to optimize the GAN objective the generator network keeps on changing even after the network can be considered optimal. As such ensembles of GANs can be constructed based on the same network initialization but just taking models which have different amount of iterations. These so-called self ensembles are much faster to train than traditional ensembles. The second method, called cascade GANs, redirects part of the training data which is badly modeled by the first GAN to another GAN. In experiments on the CIFAR10 dataset we show that ensembles of GANs obtain model probability distributions which better model the data distribution. In addition, we show that these improved results can be obtained at little additional computational cost.

[1]  Wojciech Zaremba,et al.  Improved Techniques for Training GANs , 2016, NIPS.

[2]  Trevor Darrell,et al.  Adversarial Feature Learning , 2016, ICLR.

[3]  Yoshua Bengio,et al.  Extracting and composing robust features with denoising autoencoders , 2008, ICML '08.

[4]  Yoshua Bengio,et al.  Greedy Layer-Wise Training of Deep Networks , 2006, NIPS.

[5]  Pieter Abbeel,et al.  InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets , 2016, NIPS.

[6]  Soumith Chintala,et al.  Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , 2015, ICLR.

[7]  Rob Fergus,et al.  Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks , 2015, NIPS.

[8]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[9]  S. D. Hordley,et al.  Reevaluation of color constancy algorithm performance. , 2006, Journal of the Optical Society of America. A, Optics, image science, and vision.

[10]  Matthias Bethge,et al.  A note on the evaluation of generative models , 2015, ICLR.

[11]  Hui Jiang,et al.  Generating images with recurrent adversarial networks , 2016, ArXiv.

[12]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[13]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[14]  Rob Fergus,et al.  Visualizing and Understanding Convolutional Networks , 2013, ECCV.

[15]  Pascal Vincent,et al.  Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion , 2010, J. Mach. Learn. Res..

[16]  Abhinav Gupta,et al.  Unsupervised Learning of Visual Representations Using Videos , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[17]  Nitish Srivastava.  Unsupervised Learning of Visual Representations using Videos , 2015 .