Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image...
Transcript of Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image...
![Page 1: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/1.jpg)
Generative adversarial network
![Page 2: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/2.jpg)
Analogy
![Page 3: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/3.jpg)
Analogy
![Page 4: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/4.jpg)
Analogy
![Page 5: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/5.jpg)
Analogy
adversarial
![Page 6: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/6.jpg)
Step 1: Define the problem.
Step 2: Define architecture of GAN.
Step 3: Train Discriminator on real data for n epochs
Step 4: Generate fake inputs for generator and train discriminator on fake data
Step 5: Train generator with the output of discriminator
Step 6: Repeat step 3 to step 5 for a few epochs.
Step 7: Check if the fake data manually if it seems legit. If it seems appropriate, stop training, else go to step 3.
Steps to train a GAN
![Page 7: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/7.jpg)
Natural image synthesis
![Page 8: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/8.jpg)
Application
Text to image synthesis
“inverted” CNN CNN
![Page 9: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/9.jpg)
GANs allow a model to learn that there are many correct answers
![Page 10: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/10.jpg)
Traits of GANS
Strong power to simulate
Account for uncertainty in the data
Fewer training data required
![Page 11: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/11.jpg)
Training data: vague images
Description of organism
![Page 12: Generative adversarial network - LMSElmse.org/assets/learning/workshop2019/gans.pdfText to image synthesis “inverted” CNN CNN GANs allow a model to learn that there are many correct](https://reader034.fdocuments.us/reader034/viewer/2022042319/5f09456a7e708231d426064d/html5/thumbnails/12.jpg)
Thanks