Training Generative Adversarial Networks via stochastic Nash games.

Generative adversarial networks (GANs) are a class of generative models with two antagonistic neural networks: the generator and the discriminator. These two neural networks compete against each other through an adversarial process that can be modelled as a stochastic Nash equilibrium problem. Since the associated training process is challenging, it is fundamental to design reliable algorithms to compute an equilibrium. In this paper, we propose a stochastic relaxed forward-backward algorithm for GANs and we show convergence to an exact solution or to a neighbourhood of it, if the pseudogradient mapping of the game is monotone. We apply our algorithm to the image generation problem where we observe computational advantages with respect to the extragradient scheme.

[1]  Sanja Fidler,et al.  Learning to Simulate Dynamic Environments With GameGAN , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[3]  Lei Wang,et al.  Single-Image De-Raining With Feature-Supervised Generative Adversarial Network , 2019, IEEE Signal Processing Letters.

[4]  Sergio Grammatico,et al.  Distributed Forward-Backward algorithms for stochastic generalized Nash equilibrium seeking , 2019, ArXiv.

[5]  Xiaoning Song,et al.  SP-GAN: Self-Growing and Pruning Generative Adversarial Networks , 2020, IEEE Transactions on Neural Networks and Learning Systems.

[6]  Ronald E. Bruck.  On the weak convergence of an ergodic iteration for the solution of variational inequalities for monotone operators in Hilbert space , 1977 .

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

[8]  Peter Richtárik,et al.  Revisiting Stochastic Extragradient , 2019, AISTATS.

[9]  Sebastian Nowozin,et al.  The Numerics of GANs , 2017, NIPS.

[10]  Sergio Grammatico,et al.  Comments on "Distributed robust adaptive equilibrium computation for generalized convex games" [Automatica 63 (2016) 82-91] , 2018, Autom..

[11]  Ian J. Goodfellow,et al.  NIPS 2016 Tutorial: Generative Adversarial Networks , 2016, ArXiv.

[12]  Fabio Roli,et al.  Randomized Prediction Games for Adversarial Machine Learning , 2016, IEEE Transactions on Neural Networks and Learning Systems.

[13]  Angelia Nedic,et al.  Optimal robust smoothing extragradient algorithms for stochastic variational inequality problems , 2014, 53rd IEEE Conference on Decision and Control.

[14]  Vishal M. Patel,et al.  Image De-Raining Using a Conditional Generative Adversarial Network , 2017, IEEE Transactions on Circuits and Systems for Video Technology.

[15]  Nour Eldeen M. Khalifa,et al.  Detection of Coronavirus (COVID-19) Associated Pneumonia based on Generative Adversarial Networks and a Fine-Tuned Deep Transfer Learning Model using Chest X-ray Dataset , 2020, AISI.

[16]  Alfredo N. Iusem,et al.  Extragradient Method with Variance Reduction for Stochastic Variational Inequalities , 2017, SIAM J. Optim..

[17]  Sergio Grammatico,et al.  Stochastic generalized Nash equilibrium seeking in merely monotone games , 2021, IEEE Transactions on Automatic Control.

[18]  Andreas Fischer,et al.  On generalized Nash games and variational inequalities , 2007, Oper. Res. Lett..

[19]  Sergio Grammatico,et al.  A game–theoretic approach for Generative Adversarial Networks , 2020, 2020 59th IEEE Conference on Decision and Control (CDC).

[20]  Mathias Staudigl,et al.  Minibatch Forward-Backward-Forward Methods for Solving Stochastic Variational Inequalities , 2021 .

[21]  Tomas E. Ward,et al.  Generative Adversarial Networks in Computer Vision , 2019, ACM Comput. Surv..

[22]  Paul Tseng,et al.  A Modified Forward-backward Splitting Method for Maximal Monotone Mappings 1 , 1998 .

[23]  Heinz H. Bauschke,et al.  Convex Analysis and Monotone Operator Theory in Hilbert Spaces , 2011, CMS Books in Mathematics.

[24]  Stefan Winkler,et al.  The Unusual Effectiveness of Averaging in GAN Training , 2018, ICLR.

[25]  Léon Bottou,et al.  Wasserstein GAN , 2017, ArXiv.

[26]  Phan Tu Vuong,et al.  A Relaxed Inertial Forward-Backward-Forward Algorithm for Solving Monotone Inclusions with Application to GANs , 2020, J. Mach. Learn. Res..

[27]  G. M. Korpelevich.  The extragradient method for finding saddle points and other problems , 1976 .

[28]  H. Robbins.  A Stochastic Approximation Method , 1951 .

[29]  Alex Krizhevsky,et al.  Learning Multiple Layers of Features from Tiny Images , 2009 .

[30]  Francisco Facchinei,et al.  Generalized Nash Equilibrium Problems , 2010, Ann. Oper. Res..

[31]  Gauthier Gidel,et al.  A Variational Inequality Perspective on Generative Adversarial Networks , 2018, ICLR.

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

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

[34]  F. Facchinei,et al.  Finite-Dimensional Variational Inequalities and Complementarity Problems , 2003 .

[35]  Qian Zhang,et al.  Research on image Inpainting algorithm of improved GAN based on two-discriminations networks , 2020, Applied Intelligence.

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

[37]  Uday V. Shanbhag,et al.  On the Characterization of Solution Sets of Smooth and Nonsmooth Convex Stochastic Nash Games , 2011, SIAM J. Optim..

[38]  Yura Malitsky,et al.  Golden ratio algorithms for variational inequalities , 2018, Mathematical Programming.

[39]  Zhenbing Zhao,et al.  MU-GAN: Facial Attribute Editing Based on Multi-Attention Mechanism , 2020, IEEE/CAA Journal of Automatica Sinica.

[40]  Angelia Nedic,et al.  Regularized Iterative Stochastic Approximation Methods for Stochastic Variational Inequality Problems , 2013, IEEE Transactions on Automatic Control.

[41]  Xuanqin Mou,et al.  Low-Dose CT Image Denoising Using a Generative Adversarial Network With Wasserstein Distance and Perceptual Loss , 2017, IEEE Transactions on Medical Imaging.

[42]  Chuan-Sheng Foo,et al.  Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile , 2018, ICLR.

[43]  H. Robbins,et al.  A Convergence Theorem for Non Negative Almost Supermartingales and Some Applications , 1985 .

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

[45]  Qing Tao,et al.  Stochastic Learning via Optimizing the Variational Inequalities , 2014, IEEE Transactions on Neural Networks and Learning Systems.

[46]  Fei-Yue Wang,et al.  Generative adversarial networks: introduction and outlook , 2017, IEEE/CAA Journal of Automatica Sinica.

[47]  Xi Chen,et al.  The improved image inpainting algorithm via encoder and similarity constraint , 2020, The Visual Computer.

[48]  S. Shankar Sastry,et al.  On Gradient-Based Learning in Continuous Games , 2018, SIAM J. Math. Data Sci..

[49]  Edwin D. de Jong,et al.  GANGs: Generative Adversarial Network Games , 2017, ArXiv.