Se ha denunciado esta presentación.
Utilizamos tu perfil de LinkedIn y tus datos de actividad para personalizar los anuncios y mostrarte publicidad más relevante. Puedes cambiar tus preferencias de publicidad en cualquier momento.

Introduction to GAN

4.575 visualizaciones

Publicado el

Introduction to Generative Adversarial Networks

Publicado en: Tecnología
  • Sex in your area is here: www.bit.ly/sexinarea
       Responder 
    ¿Estás seguro?    No
    Tu mensaje aparecerá aquí
  • Dating for everyone is here: www.bit.ly/2AJerkH
       Responder 
    ¿Estás seguro?    No
    Tu mensaje aparecerá aquí
  • DOWNLOAD FULL. BOOKS INTO AVAILABLE FORMAT, ......................................................................................................................... ......................................................................................................................... 1.DOWNLOAD FULL. PDF EBOOK here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. EPUB Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. doc Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. PDF EBOOK here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. EPUB Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. doc Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... ......................................................................................................................... ......................................................................................................................... .............. Browse by Genre Available eBooks ......................................................................................................................... Art, Biography, Business, Chick Lit, Children's, Christian, Classics, Comics, Contemporary, Cookbooks, Crime, Ebooks, Fantasy, Fiction, Graphic Novels, Historical Fiction, History, Horror, Humor And Comedy, Manga, Memoir, Music, Mystery, Non Fiction, Paranormal, Philosophy, Poetry, Psychology, Religion, Romance, Science, Science Fiction, Self Help, Suspense, Spirituality, Sports, Thriller, Travel, Young Adult,
       Responder 
    ¿Estás seguro?    No
    Tu mensaje aparecerá aquí
  • DOWNLOAD FULL. BOOKS INTO AVAILABLE FORMAT, ......................................................................................................................... ......................................................................................................................... 1.DOWNLOAD FULL. PDF EBOOK here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. EPUB Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. doc Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. PDF EBOOK here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. EPUB Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. doc Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... ......................................................................................................................... ......................................................................................................................... .............. Browse by Genre Available eBooks ......................................................................................................................... Art, Biography, Business, Chick Lit, Children's, Christian, Classics, Comics, Contemporary, Cookbooks, Crime, Ebooks, Fantasy, Fiction, Graphic Novels, Historical Fiction, History, Horror, Humor And Comedy, Manga, Memoir, Music, Mystery, Non Fiction, Paranormal, Philosophy, Poetry, Psychology, Religion, Romance, Science, Science Fiction, Self Help, Suspense, Spirituality, Sports, Thriller, Travel, Young Adult,
       Responder 
    ¿Estás seguro?    No
    Tu mensaje aparecerá aquí
  • DOWNLOAD FULL. BOOKS INTO AVAILABLE FORMAT, ......................................................................................................................... ......................................................................................................................... 1.DOWNLOAD FULL. PDF EBOOK here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. EPUB Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. doc Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. PDF EBOOK here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. EPUB Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... 1.DOWNLOAD FULL. doc Ebook here { https://tinyurl.com/y8nn3gmc } ......................................................................................................................... ......................................................................................................................... ......................................................................................................................... .............. Browse by Genre Available eBooks ......................................................................................................................... Art, Biography, Business, Chick Lit, Children's, Christian, Classics, Comics, Contemporary, Cookbooks, Crime, Ebooks, Fantasy, Fiction, Graphic Novels, Historical Fiction, History, Horror, Humor And Comedy, Manga, Memoir, Music, Mystery, Non Fiction, Paranormal, Philosophy, Poetry, Psychology, Religion, Romance, Science, Science Fiction, Self Help, Suspense, Spirituality, Sports, Thriller, Travel, Young Adult,
       Responder 
    ¿Estás seguro?    No
    Tu mensaje aparecerá aquí

Introduction to GAN

  1. 1. Introduction to GAN 서울대학교 방사선의학물리연구실 이 지 민 ( ljm861@gmail.com )
  2. 2. 참고 자료 출처 (본 슬라이드 인용 순) 2 좋은 자료를 만들어주신 많은 분들께 다시 한 번 감사의 인사를 전하고 싶고, 슬라이드 좌측 하단에 출처를 명시하였으니, 꼭 찾아보시길 바랍니다.  0. • • • • • • •
  3. 3. Contents 3 1. Generative Model 2. Auto-Regressive Models 3. Variational Auto-Encoder 4. Generative Adversarial Networks 5. Significant Variants & Applications
  4. 4. Generative Model 4
  5. 5. Generative Model 5 1.
  6. 6. Generative Model 6 1.
  7. 7. Generative Model 7 1.
  8. 8. Generative Model 8 1.
  9. 9. Generative Model 9 1.
  10. 10. Generative Model 10 Ideal Generative Model 1. CAT Short Hair Big Ear Model
  11. 11. Generative Model 11 Ideal Generative Model 1. CAT Short Hair Big Ear Model
  12. 12. Generative Model 12 Ideal Generative Model 1. CAT Short Hair Big Ear Tabby Model
  13. 13. Generative Model 13 Ideal Generative Model 1. CAT Short Hair Big Ear Tabby Savannah Cat Model
  14. 14. Generative Model 14 1.
  15. 15. Generative Model 15 1. https://www.slideshare.net/carpedm20/pycon-korea-2016 (지적 대화를 위한 깊고 넓은 딥러닝, 김태훈)
  16. 16. Generative Model 16 1. https://www.slideshare.net/carpedm20/pycon-korea-2016 (지적 대화를 위한 깊고 넓은 딥러닝, 김태훈)
  17. 17. Generative Model 17 1. https://www.slideshare.net/carpedm20/pycon-korea-2016 (지적 대화를 위한 깊고 넓은 딥러닝, 김태훈)
  18. 18. Generative Model 18 1. https://www.slideshare.net/carpedm20/pycon-korea-2016 (지적 대화를 위한 깊고 넓은 딥러닝, 김태훈)
  19. 19. Generative Model 19 1. https://www.slideshare.net/carpedm20/pycon-korea-2016 (지적 대화를 위한 깊고 넓은 딥러닝, 김태훈)
  20. 20. Generative Model 20 1. https://blog.openai.com/generative-models/
  21. 21. Generative Model 21 Why Generative Model 1. • • • • • • Li, Yijun, et al., Generative face completion, 2017
  22. 22. Generative Model 22 Deep Generative Models 1.  Auto-Regressive Models  Variational Auto-Encoder  Generative Adversarial Networks
  23. 23. Auto-Regressive Models 23
  24. 24. Auto-Regressive Models 24 Pixel-by-pixel generation 2. http://slazebni.cs.illinois.edu/spring17/lec13_advanced.pdf
  25. 25. Auto-Regressive Models 25 Multi-Dimensional RNNs (2013) 2. Graves et al, Multi-Dimensional Recurrent Neural Networks, 2013
  26. 26. Auto-Regressive Models 26 Spatial LSTM (2015) 2. Theis et al., Generative Image Modeling Using Spatial LSTMs, 2015
  27. 27. Auto-Regressive Models 27 Pixel RNN (2016) 2. Aaron et al, Pixel Recurrent Neural Networks, 2016
  28. 28. Auto-Regressive Models 28 Sampling 2. • ➔ • ➔ • ➔ • •
  29. 29. Auto-Regressive Models 29 Sampling 2.
  30. 30. Auto-Regressive Models 30 Sampling 2.
  31. 31. Auto-Regressive Models 31 Sampling 2.
  32. 32. Auto-Regressive Models 32 Sampling 2.
  33. 33. Auto-Regressive Models 33 Pixel RNN (2016) 2. Aaron et al, Pixel Recurrent Neural Networks, 2016
  34. 34. Auto-Regressive Models 34 Features 2. • • • •
  35. 35. Variational Auto-Encoder 35
  36. 36. Variational Auto-Encoder 36 Auto-Encoder 3. http://kvfrans.com/variational-autoencoders-explained/
  37. 37. Variational Auto-Encoder 37 Auto-Encoder 3. http://kvfrans.com/variational-autoencoders-explained/ ?
  38. 38. Variational Auto-Encoder 38 Variational Auto-Encoder 3. http://kvfrans.com/variational-autoencoders-explained/, https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/08_Autoencoder/%5B2%EA%B8%B0%5DAutoEncoder.pdf
  39. 39. Variational Auto-Encoder 39 Variational Auto-Encoder 3. http://kvfrans.com/variational-autoencoders-explained/, https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/08_Autoencoder/%5B2%EA%B8%B0%5DAutoEncoder.pdf (최건호)
  40. 40. Variational Auto-Encoder 40 Variational Auto-Encoder 3. http://kvfrans.com/variational-autoencoders-explained/, https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/08_Autoencoder/%5B2%EA%B8%B0%5DAutoEncoder.pdf (최건호)
  41. 41. Variational Auto-Encoder 41 3. http://kvfrans.com/variational-autoencoders-explained/, https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/08_Autoencoder/%5B2%EA%B8%B0%5DAutoEncoder.pdf (최건호)
  42. 42. Variational Auto-Encoder 42 3. http://kvfrans.com/variational-autoencoders-explained/, https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/08_Autoencoder/%5B2%EA%B8%B0%5DAutoEncoder.pdf (최건호)
  43. 43. Variational Auto-Encoder 43 3. http://kvfrans.com/variational-autoencoders-explained/, https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/08_Autoencoder/%5B2%EA%B8%B0%5DAutoEncoder.pdf (최건호)
  44. 44. Variational Auto-Encoder 44 3. http://kvfrans.com/variational-autoencoders-explained/, https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/08_Autoencoder/%5B2%EA%B8%B0%5DAutoEncoder.pdf (최건호)
  45. 45. Variational Auto-Encoder 45 3. http://kvfrans.com/variational-autoencoders-explained/, https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/08_Autoencoder/%5B2%EA%B8%B0%5DAutoEncoder.pdf (최건호)
  46. 46. Variational Auto-Encoder 46 Kullback-Leibler Divergence 3. https://en.wikipedia.org/wiki/Kullback–Leibler_divergence
  47. 47. Variational Auto-Encoder 47 Loss function 3. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/08_Autoencoder/%5B2%EA%B8%B0%5DAutoEncoder.pdf (최건호)
  48. 48. Variational Auto-Encoder 48 Reparameterization Trick 3. Carl Doersch, Tutorial on Variational Autoencoders, 2016
  49. 49. Variational Auto-Encoder 49 Results 3. Kingma et al., Auto-Encoding Variational Bayes, 2014
  50. 50. Variational Auto-Encoder 50 Results 3. https://www.slideshare.net/HyungjooCho2/deep-generative-modelpdf (Deep Generative Models, 조형주)
  51. 51. Variational Auto-Encoder 51 Features 3. • • • •
  52. 52. Generative Adversarial Networks 52
  53. 53. Generative Adversarial Networks 53 4.
  54. 54. Generative Adversarial Networks 54 4. 생성 모델
  55. 55. Generative Adversarial Networks 55 4. 적대적 학습
  56. 56. Generative Adversarial Networks 56 4. 적대적 학습 Generator vs Discriminator
  57. 57. Generative Adversarial Networks 57 4. https://medium.com/@devnag/generative-adversarial-networks-gans-in-50-lines-of-code-pytorch-e81b79659e3f
  58. 58. Generative Adversarial Networks 58 4. https://www.slideshare.net/ssuser77ee21/generative-adversarial-networks-70896091 (Generative Adversarial Networks, 김남주)
  59. 59. Generative Adversarial Networks 59 Value Function 4.
  60. 60. Generative Adversarial Networks 60 Value Function 4. https://www.slideshare.net/NaverEngineering/1-gangenerative-adversarial-network (1시간 만에 GAN 완전 정복하기, 최윤제)
  61. 61. Generative Adversarial Networks 61 Value Function 4. https://www.slideshare.net/NaverEngineering/1-gangenerative-adversarial-network (1시간 만에 GAN 완전 정복하기, 최윤제)
  62. 62. Generative Adversarial Networks 62 4. Goodfellow et al, Generative Adversarial Networks, 2014
  63. 63. Generative Adversarial Networks 63 Results 4. Goodfellow et al, Generative Adversarial Networks, 2014
  64. 64. Generative Adversarial Networks 64 Features 4. • • • •
  65. 65. Generative Adversarial Networks 65 Comparison with Auto-regressive models and VAE 4. http://slazebni.cs.illinois.edu/spring17/lec13_advanced.pdf, https://openai.com/blog/generative-models/
  66. 66. Generative Adversarial Networks 66 DCGAN (Deep Convolutional GAN) 4. Radford et al, Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks, 2015
  67. 67. Generative Adversarial Networks 67 DCGAN (Deep Convolutional GAN) 4. Radford et al, Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks, 2015 / https://towardsdatascience.com/activation-functions-neural-networks-1cbd9f8d91d6
  68. 68. Generative Adversarial Networks 68 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  69. 69. Generative Adversarial Networks 69 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  70. 70. Generative Adversarial Networks 70 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  71. 71. Generative Adversarial Networks 71 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  72. 72. Generative Adversarial Networks 72 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  73. 73. Generative Adversarial Networks 73 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  74. 74. Generative Adversarial Networks 74 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  75. 75. Generative Adversarial Networks 75 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  76. 76. Generative Adversarial Networks 76 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  77. 77. Generative Adversarial Networks 77 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  78. 78. Generative Adversarial Networks 78 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  79. 79. Generative Adversarial Networks 79 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  80. 80. Generative Adversarial Networks 80 Pytorch Implementation 4. https://github.com/GunhoChoi/PyTorch-FastCampus/blob/master/09_GAN/1_DCGAN/DCGAN.ipynb (최건호)
  81. 81. Generative Adversarial Networks 81 Issues during Training 4. • • • • •
  82. 82. Generative Adversarial Networks 82 Mode collapsing / Oscillating Metz et al, Unrolled Generative Adversarial Networks, 2016 4.
  83. 83. Generative Adversarial Networks 83 Mode collapsing / Oscillating 4. https://www.slideshare.net/HyungjooCho2/deep-generative-modelpdf (Deep Generative Models, 조형주)
  84. 84. Generative Adversarial Networks 84 Mode collapsing / Oscillating 4. https://www.slideshare.net/HyungjooCho2/deep-generative-modelpdf (Deep Generative Models, 조형주)
  85. 85. Generative Adversarial Networks 85 Mode collapsing / Oscillating 4. https://www.slideshare.net/HyungjooCho2/deep-generative-modelpdf (Deep Generative Models, 조형주)
  86. 86. Generative Adversarial Networks 86 Mode collapsing / Oscillating 4. Target MLE (=KL) JS Reverse KL
  87. 87. Generative Adversarial Networks 87 Mode collapsing / Oscillating 4. Metz et al, Unrolled Generative Adversarial Networks, 2016
  88. 88. Generative Adversarial Networks 88 Intractable loss 4. https://www.slideshare.net/ssuser7e10e4/wasserstein-gan-i (Wasserstein GAN 수학 이해하기, 임성빈)
  89. 89. Generative Adversarial Networks 89 Intractable loss 4. https://www.slideshare.net/ssuser7e10e4/wasserstein-gan-i (Wasserstein GAN 수학 이해하기, 임성빈)
  90. 90. Generative Adversarial Networks 90 Intractable loss 4. https://www.slideshare.net/ssuser7e10e4/wasserstein-gan-i (Wasserstein GAN 수학 이해하기, 임성빈)
  91. 91. Generative Adversarial Networks 91 Intractable loss 4. Vanilla GAN LSGAN WGAN
  92. 92. Generative Adversarial Networks 92 Intractable loss 4. Arjovsky et al, Wasserstein Generative Adversarial Networks, 2017
  93. 93. Generative Adversarial Networks 93 Intractable loss 4. Arjovsky et al, Wasserstein Generative Adversarial Networks, 2017
  94. 94. Generative Adversarial Networks 94 Balance between Generator & Discriminator 4. Berthelot et al, BEGAN, 2017
  95. 95. Generative Adversarial Networks 95 Manipulation 4. Mirza et al, Conditional Generative Adversarial Networks, 2014
  96. 96. Generative Adversarial Networks 96 Quality 4. Karras et al, Progressive Growing of GANs For Improved Quality, Stability, and Variation, 2017
  97. 97. Generative Adversarial Networks 97 Quality 4. https://www.youtube.com/watch?v=XOxxPcy5Gr4
  98. 98. Generative Adversarial Networks 98 Quality 4. Karras et al, Progressive Growing of GANs For Improved Quality, Stability, and Variation, 2017
  99. 99. Significant Variants & Applications 99
  100. 100. Significant Variants 100 Info GAN 5. Chen et al, InfoGAN, 2017
  101. 101. Significant Variants 101 Pix2Pix 5. Isola et al, Image-to-image translation with conditional GAN, 2016
  102. 102. Significant Variants 102 Domain Cross GAN 5. Taigman et al, Unsupervised Cross-Domain Image Generation, 2016
  103. 103. Significant Variants 103 CycleGAN 5. Zhu et al, Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks, 2017
  104. 104. Significant Variants 104 CycleGAN 5. Zhu et al, Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks, 2017
  105. 105. Significant Variants 105 DiscoGAN 5. Kim et al, Learning to Discover Cross Domain Relations with Generative Adversarial Networks, 2017
  106. 106. Applications 106 Time Series Generation (InfoGAN) 5. https://github.com/buriburisuri/timeseries_gan (김남주)
  107. 107. Applications 107 ehrGAN 5. Che et al., Boosting Deep Learning Risk Prediction with Generative Adversarial Networks for Electronic Health Records, 2017
  108. 108. Applications 108 RCGAN 5. Che et al., Boosting Deep Learning Risk Prediction with Generative Adversarial Networks for Electronic Health Records, 2017
  109. 109. Applications 109 GAN for Low-dose CT 5. Jelmer M. et al, Generative Adversarial Networks for Noise Reduction in Low-Dose CT, 2017, http://medicine.utah.edu/radiology/news/2016/low-dose-ct-zeng-award.php
  110. 110. Applications 110 GAN for Low-dose CT 5. Jelmer M. et al, Generative Adversarial Networks for Noise Reduction in Low-Dose CT, 2017
  111. 111. Applications 111 Stain Style Transfer 5. H Cho et al, Neural Stain-Style Transfer Learning using GAN for Histopathological Images, 2017
  112. 112. Applications 112 Simulated & Unsupervised Learning 5. Shrivastava et al, Learning from Simulated and Unsupervised Images through Adversarial Training, 2016
  113. 113. Applications 113 AnoGAN 5. Schlegl et al, Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker discovery, 2017
  114. 114. Q & A 114
  115. 115. 감사합니다. 115

×