Evaluating GANs via Duality

Generative Adversarial Networks (GANs) have shown great results in accurately modeling complex distributions, but their training is known to be difficult due to instabilities caused by a challenging minimax optimization problem. This is especially troublesome given the lack of an evaluation metric that can reliably detect non-convergent behaviors. We leverage the notion of duality gap from game theory in order to propose a novel convergence metric for GANs that has low computational cost. We verify the validity of the proposed metric for various test scenarios commonly used in the literature.

[1]  L. Deng,et al.  The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web] , 2012, IEEE Signal Processing Magazine.

[2]  Mario Lucic,et al.  Are GANs Created Equal? A Large-Scale Study , 2017, NeurIPS.

[3]  Georg Langs,et al.  Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery , 2017, IPMI.

[4]  Sepp Hochreiter,et al.  GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium , 2017, NIPS.

[5]  Max Welling,et al.  Auto-Encoding Variational Bayes , 2013, ICLR.

[6]  Bernhard Schölkopf,et al.  A Kernel Two-Sample Test , 2012, J. Mach. Learn. Res..

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

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

[9]  Sebastian Nowozin,et al.  Which Training Methods for GANs do actually Converge? , 2018, ICML.

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

[11]  Aaron C. Courville,et al.  Improved Training of Wasserstein GANs , 2017, NIPS.

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

[13]  Jaakko Lehtinen,et al.  Progressive Growing of GANs for Improved Quality, Stability, and Variation , 2017, ICLR.

[14]  J. Nash.  Equilibrium Points in N-Person Games. , 1950, Proceedings of the National Academy of Sciences of the United States of America.

[15]  Andreas Krause,et al.  An Online Learning Approach to Generative Adversarial Networks , 2017, ICLR.

[16]  Ali Borji,et al.  Pros and Cons of GAN Evaluation Measures , 2018, Comput. Vis. Image Underst..

[17]  David Pfau,et al.  Unrolled Generative Adversarial Networks , 2016, ICLR.

[18]  Ravi Kiran Sarvadevabhatla,et al.  DeLiGAN: Generative Adversarial Networks for Diverse and Limited Data , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[19]  J. Neumann.  Zur Theorie der Gesellschaftsspiele , 1928 .

[20]  Ian J. Goodfellow,et al.  Skill Rating for Generative Models , 2018, ArXiv.

[21]  Sebastian Nowozin,et al.  f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization , 2016, NIPS.

[22]  Léon Bottou,et al.  Wasserstein Generative Adversarial Networks , 2017, ICML.

[23]  Daan Wierstra,et al.  Stochastic Backpropagation and Approximate Inference in Deep Generative Models , 2014, ICML.