生成式对抗网络研究综述
|
摘要
深度学习领域一个十分活跃的分支—生成式对抗网络(GAN)已经成为人工智能学界一个热门的研究方向。生成式对抗网络采用无监督的学习方式,自动从源数据中进行学习,在不需要人工对数据集进行标记的情况下就可以产生令人惊叹的效果。阐述了GAN的背景、基本思想,对其相关理论、训练机制和应用研究进行了梳理,总结了GAN的常见网络构架、训练技巧与模型评估标准,还进行了GAN与其他生成模型VAE、衍生模型的对比,最后进行分析总结,指出GAN的优缺点并对未来发展方向进行展望。
Generative adversarial network(GAN) is an active branch of deep learning field, which has become a popular research direction in the field of artificial intelligence. GAN adopts an unsupervised learning method and automatically learns from the source data, which can produce amazing effects without artificially labeling data. In this paper, we present the background, basic idea of GAN and comb its related theory, training mechanism and state-of-the-art applications. We also summarize the common network architectures, training skills and model evaluation metrics, and compareGAN with other generative model VAE and GAN variants. Finally, we point out the advantages and disadvantages of the GAN and look forward to the further development direction.
Generative adversarial network(GAN) is an active branch of deep learning field, which has become a popular research direction in the field of artificial intelligence. GAN adopts an unsupervised learning method and automatically learns from the source data, which can produce amazing effects without artificially labeling data. In this paper, we present the background, basic idea of GAN and comb its related theory, training mechanism and state-of-the-art applications. We also summarize the common network architectures, training skills and model evaluation metrics, and compareGAN with other generative model VAE and GAN variants. Finally, we point out the advantages and disadvantages of the GAN and look forward to the further development direction.
引文
[1] DAI J,LI Y,HE K,et al.R-FCN:Object detection via region-basedfully convolutional networks [C].The Advances in Neural Information Processing Systems,2016:379-387.
[2] HONG S,ROH B,KIM K H,et al.PVANet:Lightweight deep neural networks for real-time object detection [J].Computer Vision and Pattern Recognition,2016,arXiv:1611.08588.
[3] OORD A V D,DIELEMAN S,ZEN H,et al.WaveNet:A generative model for raw audio [J].Computer Science for Sound,2016,arXiv:1609.03499.
[4] LI X,QIN T,YANG J,et al.LightRNN:Memory and computation-efficient recurrent neural networks [J].Computation and Language,2016,arXiv:1610.09893.
[5] DAUPHIN Y N,FAN A,AULI M,et al.Language modeling with gated convolutional networks [J].Computation and Language,2016,arXiv:1612.08083.
[6] 赛迪顾问.2018人工智能核心产业发展白皮书[N].中国计算机报,2018-11- 26.SAI D.White paper on core industries of Artificial Intelligencedevelopment[N].China Information World,2018-11- 26.
[7] DIEDERIK P K,MAX W.Auto-encoding variational Bayes[J].Statistics for Machine Learning,2013,ar Xiv:1312.6114.
[8] GOODFELLOW I J,POUGET-ABADIE J,MIRZA M,et al.Generative adversarial nets[C].International Conference on Neural Information Processing Systems,2014:2672- 2680.
[9] ZEN G,SANGINETO E,RICCI E,et al.Unsupervised domain adaptationfor personalized facial emotion recognition[C].International Conference on Multimodal Interaction,2014:128-135.
[10] FANNY,TJENG W C.Deep learning for imbalance data classification using class expert generative adversarial network [J].Computer Science for Machine Learning,2018,arXiv:1807.04585.
[11] WEI L,ZHIMING L,SHAOZI L.Improving deep ensemble vehicle classification by using selected adversarial samples [J].Knowledge-Based Systems,2018,160(11):167-175.
[12] BROCK A,DONAHUE J,SIMONYAN K.Large scale GAN training for high fidelity natural image synthesis[J].Computer Science for Machine Learning,2018,arXiv:1809.11096.
[13] GOODFELLOW I.NIPS 2016 tutorial:Generative adversarial networks[J].Computer Science for Machine Learning,2016 arXiv:1701.00160.
[14] 王坤峰,苟超,段艳杰,等.生成式对抗网络GAN的研究进展与展望[J].自动化学报,2017,43(3):321-332.WANG K F,GOU CH,DUAN Y J,et al.Generative adversarial networks:The State of the art and beyond[J].Acta Automatica Sinica,2017,43(3):321-332.
[15] GANGUY K.Learning Generative Adversarial Networks [M].Packt Publishing,2017.
[16] MIRZA M,SIMON O.Conditional generative adversarial nets [J].Computer Science forMachine Learning,2014,arXiv:1411.1784.
[17] ODENA A,OLAH C,SHLENS J.Conditional image synthesis with auxiliary classifier GANs[J].Statistics for Machine Learning,2016,arXiv:1610.09585.
[18] CHENG X,DUAN Y,HOUTHOOFT R,et al.Info GAN:Interpretable representation learning by information maximizing generative adversarial nets[C].Advances in Neural Information Processing Systems,2016:2172- 2180.
[19] 林懿伦,戴星原,李力,等.人工智能研究的新前线:生成式对抗网络[J].自动化学报,2018,44(5):775-792.LIN Y L,DAI X Y,LI L,et al.The new frontier of AI research:Generative adversarial networks[J].Acta Automatica Sinica,2018,44(5):775-792
[20] BENGIO Y.Learning deep architectures for AI[J].Foundations and Trends in Machine Learning,2009,2(1):1-127
[21] RADFORD A,METZ L,CHINTALA S.Unsupervised representation learning with deep convolutional generative adversarial networks[J].Computer Science forMachine Learning,2015,arXiv:1511.06434.
[22] CRESWELL A,WHITE T,DUMOULIN V,et al.Generative adversarial networks:An overview [J].IEEE Signal Processing Magazine,2017,35(1):53- 65.
[23] CRESWELL A,BHARATH A A.Inverting the generator of a generative adversarial network [J].Computer Vision and Pattern Recognition,2016,arXiv:1611.05644.
[24] LIPTON Z C,TRIPATHI S.Precise recovery of latent vectors from generative adversarial networks[J].Computer Science for Machine Learning,2017,arXiv:1702.04782.
[25] DONAHUE J,KR?HENBüHL P,DARRELL T.Adversarial feature learning[J].Computer Science for Machine Learning,2016,arXiv:1605.09782.
[26] DUMOULIN V,BELGHAZI I,POOLE B,et al.Adversarially learned inference[J].Statistics for Machine Learning,2016,arXiv:1606.00704.
[27] CHUNYUAN L,HAO L,CHANGYOU C,et al.ALICE:Towards understanding adversarial learning for joint distribution matching[J].Statistics for Machine Learning,2017,arXiv:1709.01215.
[28] THEIS L,OORD A V D,BETHGE M.A note on the evaluation of generative models[J].Statistics for Machine Learning,2015,ar Xiv:1511.01844.
[29] LARSEN A B L,LAROCHELLE H,WINTHER O.Auto encoding beyond pixels using a learned similarity metric[J].Computer Science for Machine Learning,2015,ar Xiv:1512.09300.
[30] MESCHEDER L,NOWOZIN S,GEIGER A.Adversarial variational Bayes:Unifying variational auto encoders and generative adversarial networks [J].Computer Science for Machine Learning,2017 arXiv:1701.04722.
[31] KIM T,CHA M,KIM H,et al.Learning to discover cross-domain relations with generative adversarial networks[J].Computer Vision and Pattern Recognition,2017,ar Xiv:1703.05192.
[32] ZHU J Y,PARK T,ISOLA P,et al.Unpaired image-to-image translation using cycle-consistent adversarial networks [J].Computer Vision and Pattern Recognition,2017,ar Xiv:1703.10593.
[33] YU L T,ZHANG W N,WANG J,et al.SeqGAN:Sequence generative adversarial nets with policy gradient [J].Computer Science for Machine Learning,2016,arXiv:1609.05473.
[34] SUTSKEVER I,VINYALS O,LE Q V.Sequence to sequence learning with neural networks[J].Computation and Language,2014,arXiv:1409.3215
[35] LI J,MONROE W,SHI T,et al.Adversarial learning for neural dialogue generation[J].Computation and Language,2017,arXiv:1701.06547.
[36] CHE T,LI Y,ZHANG R,et al.Maximum-Likelihood augmented discrete generative adversarial networks[J].Artificial Intelligence,2017,arXiv:1702.07983.
[37] ARJOVSKY M,BOTTOU L.Towards principled methods for training generative adversarial networks[J].Statistics for Machine Learning,2017,arXiv:1701.04862.
[38] GULRAJANI I,AHMED F,ARJOVSKY M,et al.Improved training of Wasserstein GANs[J].Computer Science for Machine Learning,2017,arXiv:1704.00028.
[39] BERTHELOT D,SCHUMM T,METZ L.BEGAN:Boundary equilibrium generative adversarial networks[J].Computer Science for Machine Learning,2017,arXiv:1703.10717.
[40] ZHAO J,MATHIEU M,LECUN Y.Energy-based generative adversarial network[J].Computer Science for Machine Learning,2016,arXiv:1609.03126.
[41] LEDIG C,WANG Z,SHI W,et al.Photo-Realistic single image super-resolution using a generative adversarial network[J].Computer Vision and Pattern Recognition,2017,arXiv:1609.04802.
[42] SALIMANS T,GOODFELLOW I,ZAREMBA W,et al.Improved techniques for training GANs[J].Computer Science for Machine Learning,2016,arXiv:1606.03498.
[43] S?NDERBY C K,CABALLERO J,THEIS L,et al.Amortised MAP inference for image super-resolution [J].Computer Vision and Pattern Recognition,2016,ar Xiv:1610.04490.
[44] IOFFE S,SZEGEDY C.Batch normalization:Accelerating deep network training by reducing internal covariate shift[J].Computer Science for Machine Learning,2015,ar Xiv:1502.03167.
[45] SPRINGENBERG J T.Unsupervised and semi-supervised learning with categorical generative adversarial networks [J].Statistics for Machine Learning,2015,arXiv:1511.06390.
[46] GOODFELLOW I J.On distinguishability criteria for estimating generative models[J].Statistics for Machine Learning,2014,ar Xiv:1412.6515.
[47] XU Q,HUANG G,YUAN Y,et al.An empirical study on evaluation metrics of generative adversarial networks[J].Computer Science for Machine Learning,2018,arXiv:1806.07755.
[48] SZEGEDY C,VANHOUCKE V,IOFFE S,et al.Rethinking the inception architecture for computer vision[J].Computer Vision and Pattern Recognition,2015,arXiv:1512.00567.
[49] CHE T,LI Y,JACBO A P,et al.Mode regularized generative adversarial networks[J].Computer Science for Machine Learning,2016,ar Xiv:1612.02136.
[50] SCH?LKOPF B,PLATT J,HOFMANN T.A kernel method for the two-sample-problem[J].Computer Science for Machine Learning,2008,arXiv:0805.2368.
[51] HEUSEL M,RAMSAUER H,UNTERTHINER T,et al.GANs trained by a two time-scale update rule converge to a local Nash equilibrium[J].Computer Science for Machine Learning,2017,arXiv:1706.08500.
[52] LOPEZ-PAZ D,OQUAB M.Revisiting classifier two-sample tests[J].Statistics for Machine Learning,2016,arXiv:1610.06545.
[53] 王坤峰,左旺孟,谭营,等.生成式对抗网络:从生成数据到创造智能[J].自动化学报,2018,44(5):769-774.WANG K F,ZUO W M,TAN Y,et al.Generative adversarial networks:from generating data to creating intelligence[J].Acta Automatica Sinica,2018,44(5):769-774.
[54] 王万良,李卓蓉.生成式对抗网络研究进展[J].通信学报,2018(2):135-148WANG W L,LI ZH R.The new frontier of AI research:Generative adversarial networks[J].Journal on Communications,2018,39(2):135-148
[55] KARRAS T,AILA T,LAINE S,et al.Progressive growing of GANs for improved quality,stability,and variation[J].Neural and Evolutionary Computing,2017,arXiv:1710.10196.
[56] WANG Y,PERAZZI F,MCWILLIAMS B,et al.A fully progressive approach to single-image super-resolution [J].Computer Vision and Pattern Recognition,2018,arXiv:1804.02900.
[57] JIN Y,ZHANG J,LI M,et al.Towards the automatic anime characters creation with generative adversarial networks[J].Computer Vision and Pattern Recognition,2017,arXiv:1708.05509.
[58] ISOLA P,ZHU J Y,ZHOU T H,et al.Image-to-image translation with conditional adversarial networks[J].Computer Vision and Pattern Recognition,2016,arXiv:1611.07004.
[59] WANG T C,LIU M Y,ZHU J Y,et al.High-resolution image synthesis and semantic manipulation with conditional GANs[J].Computer Vision and Pattern Recognition,2017,arXiv:1711.11585.
[60] CHOI Y,CHOI M,KIM M,et al.StarGAN:Unified generative adversarial networks for multi- domain image-to-image translation [J].Computer Vision and Pattern Recognition,2017,arXiv:1711.09020.
[61] YOO D,KIM N,PARK S,et al.Pixel-level domain transfer[J].Computer Vision and Pattern Recognition,2016,arXiv:1603.07442.
[62] HUANG R,ZHANG S,LI T,et al.Beyond face rotation:Global and local perception GAN for photorealistic and identity preserving frontal view synthesis[C].IEEE International Conference on Computer Vision,2017:2458- 2467.
[63] SANTANA E,HOTZ G.Learning a driving simulator[J].Computer Science for Machine Learning,2016,arXiv:1608.01230.
[64] LIU S,SUN Y,ZHU D,et al.Face aging with contextual generative adversarial nets[J].Computer Vision and Pattern Recognition,2017,arXiv:1702.01983.
[65] LIU G,REDA F A,SHIH K J,et al.Image inpainting for irregular holes using partial convolutions[J].Computer Vision and Pattern Recognition,2018,arXiv:1804.07723.
[66] VONDRICK C,PIRSIAVASH H,TORRALBA A.Generating videos with scene dynamics[J].Computer Vision and Pattern Recognition,2016,arXiv:1609.02612.
[67] HUANG X,LI Y,POURSAEED O,et al.Stacked generative adversarial networks[J].Computer Vision and Pattern Recognition,2016,arXiv:1612.04357.
[68] ZHANG H,XU T,LI H,et al.Stack GAN:Text to photo-realistic image synthesis with stacked generative adversarial networks[J].Computer Vision and Pattern Recognition,2016,ar Xiv:1612.03242.
[69] XU T,ZHANG P,HUANG Q,et al.AttnGAN:Fine-grained text to image generation with attentional generative adversarial networks[J].Computer Vision and Pattern Recognition,2017,arXiv:1711.10485.
[70] LIU B,FU J,KATO M P,et al.Beyond narrative description:Generating poetry from images by multi-adversarial training[J].Computer Vision and Pattern Recognition,2018,arXiv:1804.08473.
[71] FEDUS W,GOODFELLOW I,DAI A M.MaskGAN:better text generation via filling in the__[J].Statistics for Machine Learning,2018,arXiv:1801.07736.
[72] BRUNNER G,WANG Y,WATTENHOFER R,et al.Symbolic music genre transfer with CycleGAN[J].Computer Science for Sound,2018,arXiv:1809.07575.
[73] DONG H W,HSIAO W Y,YANG L C,et al.MuseGAN:Multi-track sequential generative adversarial networks for symbolic music generation and accompaniment[J].Audio and Speech Processing,2017,arXiv:1709.06298.
[74] WU J,ZHANG C,XUE T,et al.Learning a probabilistic latent space of object shapes via 3D generative-adversarial modeling[J].Computer Vision and Pattern Recognition,2017,arXiv:1610.07584.
[75] YONG O L,JO J,HWANG J.Application of deep neural network and generative adversarial network to industrial maintenance:A case study of induction motor fault detection[C].IEEE International Conference on Big Data,2018:3248-3253.
[76] SCHLEGL T,SEEB?CK P,WALDSTEIN S M,Unsupervised anomaly detection with generative adversarial networks to guide marker discovery[J].Computer Vision and Pattern Recognition,2017,arXiv:1703.05921.
[2] HONG S,ROH B,KIM K H,et al.PVANet:Lightweight deep neural networks for real-time object detection [J].Computer Vision and Pattern Recognition,2016,arXiv:1611.08588.
[3] OORD A V D,DIELEMAN S,ZEN H,et al.WaveNet:A generative model for raw audio [J].Computer Science for Sound,2016,arXiv:1609.03499.
[4] LI X,QIN T,YANG J,et al.LightRNN:Memory and computation-efficient recurrent neural networks [J].Computation and Language,2016,arXiv:1610.09893.
[5] DAUPHIN Y N,FAN A,AULI M,et al.Language modeling with gated convolutional networks [J].Computation and Language,2016,arXiv:1612.08083.
[6] 赛迪顾问.2018人工智能核心产业发展白皮书[N].中国计算机报,2018-11- 26.SAI D.White paper on core industries of Artificial Intelligencedevelopment[N].China Information World,2018-11- 26.
[7] DIEDERIK P K,MAX W.Auto-encoding variational Bayes[J].Statistics for Machine Learning,2013,ar Xiv:1312.6114.
[8] GOODFELLOW I J,POUGET-ABADIE J,MIRZA M,et al.Generative adversarial nets[C].International Conference on Neural Information Processing Systems,2014:2672- 2680.
[9] ZEN G,SANGINETO E,RICCI E,et al.Unsupervised domain adaptationfor personalized facial emotion recognition[C].International Conference on Multimodal Interaction,2014:128-135.
[10] FANNY,TJENG W C.Deep learning for imbalance data classification using class expert generative adversarial network [J].Computer Science for Machine Learning,2018,arXiv:1807.04585.
[11] WEI L,ZHIMING L,SHAOZI L.Improving deep ensemble vehicle classification by using selected adversarial samples [J].Knowledge-Based Systems,2018,160(11):167-175.
[12] BROCK A,DONAHUE J,SIMONYAN K.Large scale GAN training for high fidelity natural image synthesis[J].Computer Science for Machine Learning,2018,arXiv:1809.11096.
[13] GOODFELLOW I.NIPS 2016 tutorial:Generative adversarial networks[J].Computer Science for Machine Learning,2016 arXiv:1701.00160.
[14] 王坤峰,苟超,段艳杰,等.生成式对抗网络GAN的研究进展与展望[J].自动化学报,2017,43(3):321-332.WANG K F,GOU CH,DUAN Y J,et al.Generative adversarial networks:The State of the art and beyond[J].Acta Automatica Sinica,2017,43(3):321-332.
[15] GANGUY K.Learning Generative Adversarial Networks [M].Packt Publishing,2017.
[16] MIRZA M,SIMON O.Conditional generative adversarial nets [J].Computer Science forMachine Learning,2014,arXiv:1411.1784.
[17] ODENA A,OLAH C,SHLENS J.Conditional image synthesis with auxiliary classifier GANs[J].Statistics for Machine Learning,2016,arXiv:1610.09585.
[18] CHENG X,DUAN Y,HOUTHOOFT R,et al.Info GAN:Interpretable representation learning by information maximizing generative adversarial nets[C].Advances in Neural Information Processing Systems,2016:2172- 2180.
[19] 林懿伦,戴星原,李力,等.人工智能研究的新前线:生成式对抗网络[J].自动化学报,2018,44(5):775-792.LIN Y L,DAI X Y,LI L,et al.The new frontier of AI research:Generative adversarial networks[J].Acta Automatica Sinica,2018,44(5):775-792
[20] BENGIO Y.Learning deep architectures for AI[J].Foundations and Trends in Machine Learning,2009,2(1):1-127
[21] RADFORD A,METZ L,CHINTALA S.Unsupervised representation learning with deep convolutional generative adversarial networks[J].Computer Science forMachine Learning,2015,arXiv:1511.06434.
[22] CRESWELL A,WHITE T,DUMOULIN V,et al.Generative adversarial networks:An overview [J].IEEE Signal Processing Magazine,2017,35(1):53- 65.
[23] CRESWELL A,BHARATH A A.Inverting the generator of a generative adversarial network [J].Computer Vision and Pattern Recognition,2016,arXiv:1611.05644.
[24] LIPTON Z C,TRIPATHI S.Precise recovery of latent vectors from generative adversarial networks[J].Computer Science for Machine Learning,2017,arXiv:1702.04782.
[25] DONAHUE J,KR?HENBüHL P,DARRELL T.Adversarial feature learning[J].Computer Science for Machine Learning,2016,arXiv:1605.09782.
[26] DUMOULIN V,BELGHAZI I,POOLE B,et al.Adversarially learned inference[J].Statistics for Machine Learning,2016,arXiv:1606.00704.
[27] CHUNYUAN L,HAO L,CHANGYOU C,et al.ALICE:Towards understanding adversarial learning for joint distribution matching[J].Statistics for Machine Learning,2017,arXiv:1709.01215.
[28] THEIS L,OORD A V D,BETHGE M.A note on the evaluation of generative models[J].Statistics for Machine Learning,2015,ar Xiv:1511.01844.
[29] LARSEN A B L,LAROCHELLE H,WINTHER O.Auto encoding beyond pixels using a learned similarity metric[J].Computer Science for Machine Learning,2015,ar Xiv:1512.09300.
[30] MESCHEDER L,NOWOZIN S,GEIGER A.Adversarial variational Bayes:Unifying variational auto encoders and generative adversarial networks [J].Computer Science for Machine Learning,2017 arXiv:1701.04722.
[31] KIM T,CHA M,KIM H,et al.Learning to discover cross-domain relations with generative adversarial networks[J].Computer Vision and Pattern Recognition,2017,ar Xiv:1703.05192.
[32] ZHU J Y,PARK T,ISOLA P,et al.Unpaired image-to-image translation using cycle-consistent adversarial networks [J].Computer Vision and Pattern Recognition,2017,ar Xiv:1703.10593.
[33] YU L T,ZHANG W N,WANG J,et al.SeqGAN:Sequence generative adversarial nets with policy gradient [J].Computer Science for Machine Learning,2016,arXiv:1609.05473.
[34] SUTSKEVER I,VINYALS O,LE Q V.Sequence to sequence learning with neural networks[J].Computation and Language,2014,arXiv:1409.3215
[35] LI J,MONROE W,SHI T,et al.Adversarial learning for neural dialogue generation[J].Computation and Language,2017,arXiv:1701.06547.
[36] CHE T,LI Y,ZHANG R,et al.Maximum-Likelihood augmented discrete generative adversarial networks[J].Artificial Intelligence,2017,arXiv:1702.07983.
[37] ARJOVSKY M,BOTTOU L.Towards principled methods for training generative adversarial networks[J].Statistics for Machine Learning,2017,arXiv:1701.04862.
[38] GULRAJANI I,AHMED F,ARJOVSKY M,et al.Improved training of Wasserstein GANs[J].Computer Science for Machine Learning,2017,arXiv:1704.00028.
[39] BERTHELOT D,SCHUMM T,METZ L.BEGAN:Boundary equilibrium generative adversarial networks[J].Computer Science for Machine Learning,2017,arXiv:1703.10717.
[40] ZHAO J,MATHIEU M,LECUN Y.Energy-based generative adversarial network[J].Computer Science for Machine Learning,2016,arXiv:1609.03126.
[41] LEDIG C,WANG Z,SHI W,et al.Photo-Realistic single image super-resolution using a generative adversarial network[J].Computer Vision and Pattern Recognition,2017,arXiv:1609.04802.
[42] SALIMANS T,GOODFELLOW I,ZAREMBA W,et al.Improved techniques for training GANs[J].Computer Science for Machine Learning,2016,arXiv:1606.03498.
[43] S?NDERBY C K,CABALLERO J,THEIS L,et al.Amortised MAP inference for image super-resolution [J].Computer Vision and Pattern Recognition,2016,ar Xiv:1610.04490.
[44] IOFFE S,SZEGEDY C.Batch normalization:Accelerating deep network training by reducing internal covariate shift[J].Computer Science for Machine Learning,2015,ar Xiv:1502.03167.
[45] SPRINGENBERG J T.Unsupervised and semi-supervised learning with categorical generative adversarial networks [J].Statistics for Machine Learning,2015,arXiv:1511.06390.
[46] GOODFELLOW I J.On distinguishability criteria for estimating generative models[J].Statistics for Machine Learning,2014,ar Xiv:1412.6515.
[47] XU Q,HUANG G,YUAN Y,et al.An empirical study on evaluation metrics of generative adversarial networks[J].Computer Science for Machine Learning,2018,arXiv:1806.07755.
[48] SZEGEDY C,VANHOUCKE V,IOFFE S,et al.Rethinking the inception architecture for computer vision[J].Computer Vision and Pattern Recognition,2015,arXiv:1512.00567.
[49] CHE T,LI Y,JACBO A P,et al.Mode regularized generative adversarial networks[J].Computer Science for Machine Learning,2016,ar Xiv:1612.02136.
[50] SCH?LKOPF B,PLATT J,HOFMANN T.A kernel method for the two-sample-problem[J].Computer Science for Machine Learning,2008,arXiv:0805.2368.
[51] HEUSEL M,RAMSAUER H,UNTERTHINER T,et al.GANs trained by a two time-scale update rule converge to a local Nash equilibrium[J].Computer Science for Machine Learning,2017,arXiv:1706.08500.
[52] LOPEZ-PAZ D,OQUAB M.Revisiting classifier two-sample tests[J].Statistics for Machine Learning,2016,arXiv:1610.06545.
[53] 王坤峰,左旺孟,谭营,等.生成式对抗网络:从生成数据到创造智能[J].自动化学报,2018,44(5):769-774.WANG K F,ZUO W M,TAN Y,et al.Generative adversarial networks:from generating data to creating intelligence[J].Acta Automatica Sinica,2018,44(5):769-774.
[54] 王万良,李卓蓉.生成式对抗网络研究进展[J].通信学报,2018(2):135-148WANG W L,LI ZH R.The new frontier of AI research:Generative adversarial networks[J].Journal on Communications,2018,39(2):135-148
[55] KARRAS T,AILA T,LAINE S,et al.Progressive growing of GANs for improved quality,stability,and variation[J].Neural and Evolutionary Computing,2017,arXiv:1710.10196.
[56] WANG Y,PERAZZI F,MCWILLIAMS B,et al.A fully progressive approach to single-image super-resolution [J].Computer Vision and Pattern Recognition,2018,arXiv:1804.02900.
[57] JIN Y,ZHANG J,LI M,et al.Towards the automatic anime characters creation with generative adversarial networks[J].Computer Vision and Pattern Recognition,2017,arXiv:1708.05509.
[58] ISOLA P,ZHU J Y,ZHOU T H,et al.Image-to-image translation with conditional adversarial networks[J].Computer Vision and Pattern Recognition,2016,arXiv:1611.07004.
[59] WANG T C,LIU M Y,ZHU J Y,et al.High-resolution image synthesis and semantic manipulation with conditional GANs[J].Computer Vision and Pattern Recognition,2017,arXiv:1711.11585.
[60] CHOI Y,CHOI M,KIM M,et al.StarGAN:Unified generative adversarial networks for multi- domain image-to-image translation [J].Computer Vision and Pattern Recognition,2017,arXiv:1711.09020.
[61] YOO D,KIM N,PARK S,et al.Pixel-level domain transfer[J].Computer Vision and Pattern Recognition,2016,arXiv:1603.07442.
[62] HUANG R,ZHANG S,LI T,et al.Beyond face rotation:Global and local perception GAN for photorealistic and identity preserving frontal view synthesis[C].IEEE International Conference on Computer Vision,2017:2458- 2467.
[63] SANTANA E,HOTZ G.Learning a driving simulator[J].Computer Science for Machine Learning,2016,arXiv:1608.01230.
[64] LIU S,SUN Y,ZHU D,et al.Face aging with contextual generative adversarial nets[J].Computer Vision and Pattern Recognition,2017,arXiv:1702.01983.
[65] LIU G,REDA F A,SHIH K J,et al.Image inpainting for irregular holes using partial convolutions[J].Computer Vision and Pattern Recognition,2018,arXiv:1804.07723.
[66] VONDRICK C,PIRSIAVASH H,TORRALBA A.Generating videos with scene dynamics[J].Computer Vision and Pattern Recognition,2016,arXiv:1609.02612.
[67] HUANG X,LI Y,POURSAEED O,et al.Stacked generative adversarial networks[J].Computer Vision and Pattern Recognition,2016,arXiv:1612.04357.
[68] ZHANG H,XU T,LI H,et al.Stack GAN:Text to photo-realistic image synthesis with stacked generative adversarial networks[J].Computer Vision and Pattern Recognition,2016,ar Xiv:1612.03242.
[69] XU T,ZHANG P,HUANG Q,et al.AttnGAN:Fine-grained text to image generation with attentional generative adversarial networks[J].Computer Vision and Pattern Recognition,2017,arXiv:1711.10485.
[70] LIU B,FU J,KATO M P,et al.Beyond narrative description:Generating poetry from images by multi-adversarial training[J].Computer Vision and Pattern Recognition,2018,arXiv:1804.08473.
[71] FEDUS W,GOODFELLOW I,DAI A M.MaskGAN:better text generation via filling in the__[J].Statistics for Machine Learning,2018,arXiv:1801.07736.
[72] BRUNNER G,WANG Y,WATTENHOFER R,et al.Symbolic music genre transfer with CycleGAN[J].Computer Science for Sound,2018,arXiv:1809.07575.
[73] DONG H W,HSIAO W Y,YANG L C,et al.MuseGAN:Multi-track sequential generative adversarial networks for symbolic music generation and accompaniment[J].Audio and Speech Processing,2017,arXiv:1709.06298.
[74] WU J,ZHANG C,XUE T,et al.Learning a probabilistic latent space of object shapes via 3D generative-adversarial modeling[J].Computer Vision and Pattern Recognition,2017,arXiv:1610.07584.
[75] YONG O L,JO J,HWANG J.Application of deep neural network and generative adversarial network to industrial maintenance:A case study of induction motor fault detection[C].IEEE International Conference on Big Data,2018:3248-3253.
[76] SCHLEGL T,SEEB?CK P,WALDSTEIN S M,Unsupervised anomaly detection with generative adversarial networks to guide marker discovery[J].Computer Vision and Pattern Recognition,2017,arXiv:1703.05921.