SlideShare a Scribd company logo
1 of 42
Download to read offline
Generative adversarial networks
and their applications
by Vijay Veerabadran, Research Engineer (CV) , Artifacia
(@vijayvee)
January 7, 2017
Meet agenda
● Phase 1 : Introduction to generative models and GANs
● Phase 2 : Types of GANs
● Phase 3 : Applications of GANs
● Phase 4 : Limitations of GANs
● Phase 5 : GAN Hacks
Generative models
● A model that can generate seemingly natural data
samples
Generative
adversarial networks
Analogy to counterfeit
currency
Generator Discriminator
Motivation
● Necessity is the mother of invention
● Previous approaches used for modeling were based on
Maximum Likelihood Estimation, Variational Autoencoders
etc.
● Huge intractable probabilistic computations
● Need for a more tractable and easily trainable model
leads to the inception of GANs
Generative Adversarial Nets - Ian et al
Noise
Generative Adversarial Nets - Ian et al
Training stages of a GAN
Black dotted: True data
Green solid: Generated data
Blue dotted: Discriminator loss
● Loss function proposed in Goodfellow’s paper introducing
GANs
Adversarial loss
True data Noise provided for
generating data
Bonus! - Vector space
arithmetic of GANs
Types of GANs
Sub classifications
Basic vanilla GAN (T1)
Deep Convolutional GAN
(T1)
Conditional GAN (T2)
Bidirectional GAN (T2)
Semi-supervised GAN (T3)
Applications of
GANs
Image generation
PPGNs - High resolution image generation
Interactive image generation - iGANs
GANs for Diagrammatic Abstract
Reasoning
Image super resolution
using SRGANs
(Result of our experiment
at Artifacia)
Image inpainting
Input to GAN GAN generated
output
(Result of our experiment
at Artifacia)
Semantic segmentation using
Adversarial networks
Video generation using
Adversarial networks
Generator network
Impressive text to image results
Limitations of
GANs
Mode collapse
● Generator keeps generating highly similar looking
images
● Happens when the generator is optimized while
keeping the discriminator constant for many
iterations
Predicting pixels based on context
● GANs are trained to predict all pixels in an image
at once
● Giving one pixel and predicting its neighbouring
pixels is hence difficult
Difficulty in reaching convergence
● Generator and discriminator losses keep oscillating
● Network does not converge to an optimal solution
Relative strength of the two
networks
● Either of the two networks becoming extremely
strong relative to the other
● Network never learns beyond this point
Dealing with these issues
- GAN Hacks
Normalizing images
● Standard practice of normalizing images by mean
normalizing and scaling by stddev should work
● Make sure that the images are normalized to
values between -1 and +1
● Paper explaining intuition Sampling Generative
Networks
Inverting labels
● While training generator, flip labels sent to the
discriminator
● Label of real image is made fake(say class 0) and
that of generated image is made real(class 1)
Schedule for training G and D
● People tend to find a fixed (nIter_Gen/nIter_Disc)
training schedule
● Better to train G or D based on a loss threshold
● if loss_G > A:
train_G()
If loss_D > B:
train_D()
Noisy D input
● Add noise to inputs of the discriminator
● Label of real image is made fake(say class 0) and
that of generated image is made real(class 1)
● Paper explaining intuition: Towards Principled
Methods for Training Generative Adversarial
Networks
Feature matching
● Modify the loss function of the generator to include
intermediate feature activations rather than output
● Forces the generator to generate data that match
the statistics of real data
One sided label smoothing
● Technique proposed in the 1980’s
● Instead of having hard labels like 0 and 1,
smoothen the labels by making them close to 0
and 1
● For example, 0,1 -> 0.1,0.9
References
● Ian's talk on GANs
● Generative Adversarial Networks - Ian et al
● DCGANs
● Conditional GANs
● BiGANs
● Semi-supervised GANs
● PPGN
● iGANs
● Text to image generation
● Improved techniques for training GANs
● GAN Hacks
“What I cannot create, I do not understand”
- Richard Feynman
THANK YOU
Join
meetup.com/Artifacia-AI-Meet/

More Related Content

What's hot

Deep learning - A Visual Introduction
Deep learning - A Visual IntroductionDeep learning - A Visual Introduction
Deep learning - A Visual Introduction
Lukas Masuch
 
Image-to-Image Translation with Conditional Adversarial Nets (UPC Reading Group)
Image-to-Image Translation with Conditional Adversarial Nets (UPC Reading Group)Image-to-Image Translation with Conditional Adversarial Nets (UPC Reading Group)
Image-to-Image Translation with Conditional Adversarial Nets (UPC Reading Group)
Universitat Politècnica de Catalunya
 

What's hot (20)

Generative Adversarial Networks (GANs)
Generative Adversarial Networks (GANs)Generative Adversarial Networks (GANs)
Generative Adversarial Networks (GANs)
 
Generative adversarial networks
Generative adversarial networksGenerative adversarial networks
Generative adversarial networks
 
Introduction to Generative Adversarial Networks (GANs)
Introduction to Generative Adversarial Networks (GANs)Introduction to Generative Adversarial Networks (GANs)
Introduction to Generative Adversarial Networks (GANs)
 
Deep Learning for Computer Vision: Generative models and adversarial training...
Deep Learning for Computer Vision: Generative models and adversarial training...Deep Learning for Computer Vision: Generative models and adversarial training...
Deep Learning for Computer Vision: Generative models and adversarial training...
 
GANs and Applications
GANs and ApplicationsGANs and Applications
GANs and Applications
 
GANs Deep Learning Summer School
GANs Deep Learning Summer SchoolGANs Deep Learning Summer School
GANs Deep Learning Summer School
 
GAN - Theory and Applications
GAN - Theory and ApplicationsGAN - Theory and Applications
GAN - Theory and Applications
 
Deep learning - A Visual Introduction
Deep learning - A Visual IntroductionDeep learning - A Visual Introduction
Deep learning - A Visual Introduction
 
A Short Introduction to Generative Adversarial Networks
A Short Introduction to Generative Adversarial NetworksA Short Introduction to Generative Adversarial Networks
A Short Introduction to Generative Adversarial Networks
 
Gan intro
Gan introGan intro
Gan intro
 
Generative Adversarial Network (GAN)
Generative Adversarial Network (GAN)Generative Adversarial Network (GAN)
Generative Adversarial Network (GAN)
 
Introduction to Generative Adversarial Networks (GAN) with Apache MXNet
Introduction to Generative Adversarial Networks (GAN) with Apache MXNetIntroduction to Generative Adversarial Networks (GAN) with Apache MXNet
Introduction to Generative Adversarial Networks (GAN) with Apache MXNet
 
Transfer Learning
Transfer LearningTransfer Learning
Transfer Learning
 
Anomaly Detection Using Generative Adversarial Network(GAN)
Anomaly Detection Using Generative Adversarial Network(GAN)Anomaly Detection Using Generative Adversarial Network(GAN)
Anomaly Detection Using Generative Adversarial Network(GAN)
 
Introduction to Generative Adversarial Networks
Introduction to Generative Adversarial NetworksIntroduction to Generative Adversarial Networks
Introduction to Generative Adversarial Networks
 
GANs Presentation.pptx
GANs Presentation.pptxGANs Presentation.pptx
GANs Presentation.pptx
 
Journey of Generative AI
Journey of Generative AIJourney of Generative AI
Journey of Generative AI
 
Image-to-Image Translation with Conditional Adversarial Nets (UPC Reading Group)
Image-to-Image Translation with Conditional Adversarial Nets (UPC Reading Group)Image-to-Image Translation with Conditional Adversarial Nets (UPC Reading Group)
Image-to-Image Translation with Conditional Adversarial Nets (UPC Reading Group)
 
Generative Adversarial Networks and Their Medical Imaging Applications
Generative Adversarial Networks and Their Medical Imaging ApplicationsGenerative Adversarial Networks and Their Medical Imaging Applications
Generative Adversarial Networks and Their Medical Imaging Applications
 
Deep learning for real life applications
Deep learning for real life applicationsDeep learning for real life applications
Deep learning for real life applications
 

Viewers also liked

Exploring temporal graph data with Python: 
a study on tensor decomposition o...
Exploring temporal graph data with Python: 
a study on tensor decomposition o...Exploring temporal graph data with Python: 
a study on tensor decomposition o...
Exploring temporal graph data with Python: 
a study on tensor decomposition o...
André Panisson
 
Zeromq - Pycon India 2013
Zeromq - Pycon India 2013Zeromq - Pycon India 2013
Zeromq - Pycon India 2013
Srinivasan R
 

Viewers also liked (10)

Exploring temporal graph data with Python: 
a study on tensor decomposition o...
Exploring temporal graph data with Python: 
a study on tensor decomposition o...Exploring temporal graph data with Python: 
a study on tensor decomposition o...
Exploring temporal graph data with Python: 
a study on tensor decomposition o...
 
Principal Component Analysis for Tensor Analysis and EEG classification
Principal Component Analysis for Tensor Analysis and EEG classificationPrincipal Component Analysis for Tensor Analysis and EEG classification
Principal Component Analysis for Tensor Analysis and EEG classification
 
[DL輪読会]Image-to-Image Translation with Conditional Adversarial Networks
[DL輪読会]Image-to-Image Translation with Conditional Adversarial Networks[DL輪読会]Image-to-Image Translation with Conditional Adversarial Networks
[DL輪読会]Image-to-Image Translation with Conditional Adversarial Networks
 
Machine learning Lecture 2
Machine learning Lecture 2Machine learning Lecture 2
Machine learning Lecture 2
 
Machine learning Lecture 3
Machine learning Lecture 3Machine learning Lecture 3
Machine learning Lecture 3
 
Zeromq - Pycon India 2013
Zeromq - Pycon India 2013Zeromq - Pycon India 2013
Zeromq - Pycon India 2013
 
Machine learning Lecture 1
Machine learning Lecture 1Machine learning Lecture 1
Machine learning Lecture 1
 
Candidate elimination example
Candidate elimination exampleCandidate elimination example
Candidate elimination example
 
Machine learning Lecture 4
Machine learning Lecture 4Machine learning Lecture 4
Machine learning Lecture 4
 
Machine Learning Lecture 3 Decision Trees
Machine Learning Lecture 3 Decision TreesMachine Learning Lecture 3 Decision Trees
Machine Learning Lecture 3 Decision Trees
 

Similar to Generative Adversarial Networks and Their Applications

Jakub Langr (University of Oxford) - Overview of Generative Adversarial Netwo...
Jakub Langr (University of Oxford) - Overview of Generative Adversarial Netwo...Jakub Langr (University of Oxford) - Overview of Generative Adversarial Netwo...
Jakub Langr (University of Oxford) - Overview of Generative Adversarial Netwo...
Codiax
 
GAN Deep Learning Approaches to Image Processing Applications (1).pptx
GAN Deep Learning Approaches to Image Processing Applications (1).pptxGAN Deep Learning Approaches to Image Processing Applications (1).pptx
GAN Deep Learning Approaches to Image Processing Applications (1).pptx
RMDAcademicCoordinat
 
GNA 13552928 deep learning for GAN a.ppt
GNA 13552928 deep learning for GAN a.pptGNA 13552928 deep learning for GAN a.ppt
GNA 13552928 deep learning for GAN a.ppt
ManiMaran230751
 

Similar to Generative Adversarial Networks and Their Applications (20)

Deep Generative Modelling
Deep Generative ModellingDeep Generative Modelling
Deep Generative Modelling
 
Generative Adversarial Networks and Their Applications in Medical Imaging
Generative Adversarial Networks  and Their Applications in Medical ImagingGenerative Adversarial Networks  and Their Applications in Medical Imaging
Generative Adversarial Networks and Their Applications in Medical Imaging
 
brief Introduction to Different Kinds of GANs
brief Introduction to Different Kinds of GANsbrief Introduction to Different Kinds of GANs
brief Introduction to Different Kinds of GANs
 
Jakub Langr (University of Oxford) - Overview of Generative Adversarial Netwo...
Jakub Langr (University of Oxford) - Overview of Generative Adversarial Netwo...Jakub Langr (University of Oxford) - Overview of Generative Adversarial Netwo...
Jakub Langr (University of Oxford) - Overview of Generative Adversarial Netwo...
 
GAN Deep Learning Approaches to Image Processing Applications (1).pptx
GAN Deep Learning Approaches to Image Processing Applications (1).pptxGAN Deep Learning Approaches to Image Processing Applications (1).pptx
GAN Deep Learning Approaches to Image Processing Applications (1).pptx
 
GNA 13552928 deep learning for GAN a.ppt
GNA 13552928 deep learning for GAN a.pptGNA 13552928 deep learning for GAN a.ppt
GNA 13552928 deep learning for GAN a.ppt
 
gan.pdf
gan.pdfgan.pdf
gan.pdf
 
Google Dev Group Yangon (2020) AI Talk (Creative AI in Action)
Google Dev Group Yangon (2020)  AI Talk (Creative AI in Action)Google Dev Group Yangon (2020)  AI Talk (Creative AI in Action)
Google Dev Group Yangon (2020) AI Talk (Creative AI in Action)
 
Malenko Аndrii "Security for AI"
Malenko Аndrii "Security for AI"Malenko Аndrii "Security for AI"
Malenko Аndrii "Security for AI"
 
Generative modeling with Convolutional Neural Networks
Generative modeling with Convolutional Neural NetworksGenerative modeling with Convolutional Neural Networks
Generative modeling with Convolutional Neural Networks
 
Reading group gan - 20170417
Reading group   gan - 20170417Reading group   gan - 20170417
Reading group gan - 20170417
 
Exploring Generative AI with GAN Models
Exploring Generative AI with GAN ModelsExploring Generative AI with GAN Models
Exploring Generative AI with GAN Models
 
Nips 2016 tutorial generative adversarial networks review
Nips 2016 tutorial  generative adversarial networks reviewNips 2016 tutorial  generative adversarial networks review
Nips 2016 tutorial generative adversarial networks review
 
Generative adversarial network_Ayadi_Alaeddine
Generative adversarial network_Ayadi_AlaeddineGenerative adversarial network_Ayadi_Alaeddine
Generative adversarial network_Ayadi_Alaeddine
 
Deep Generative Models - Kevin McGuinness - UPC Barcelona 2018
Deep Generative Models - Kevin McGuinness - UPC Barcelona 2018Deep Generative Models - Kevin McGuinness - UPC Barcelona 2018
Deep Generative Models - Kevin McGuinness - UPC Barcelona 2018
 
EuroSciPy 2019 - GANs: Theory and Applications
EuroSciPy 2019 - GANs: Theory and ApplicationsEuroSciPy 2019 - GANs: Theory and Applications
EuroSciPy 2019 - GANs: Theory and Applications
 
IRJET- Generating 3D Models Using 3D Generative Adversarial Network
IRJET- Generating 3D Models Using 3D Generative Adversarial NetworkIRJET- Generating 3D Models Using 3D Generative Adversarial Network
IRJET- Generating 3D Models Using 3D Generative Adversarial Network
 
ExplainingMLModels.pdf
ExplainingMLModels.pdfExplainingMLModels.pdf
ExplainingMLModels.pdf
 
Generative advesarial networks technical seminar
Generative advesarial networks technical seminarGenerative advesarial networks technical seminar
Generative advesarial networks technical seminar
 
An Extensive Review on Generative Adversarial Networks GAN’s
An Extensive Review on Generative Adversarial Networks GAN’sAn Extensive Review on Generative Adversarial Networks GAN’s
An Extensive Review on Generative Adversarial Networks GAN’s
 

Recently uploaded

Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
WSO2
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Safe Software
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Safe Software
 

Recently uploaded (20)

AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Ransomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfRansomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdf
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...
 
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In PakistanCNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistan
 
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamDEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
 
Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
 
ICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesICT role in 21st century education and its challenges
ICT role in 21st century education and its challenges
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century education
 
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
 
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfRising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
 
Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyArtificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : Uncertainty
 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptx
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
 

Generative Adversarial Networks and Their Applications