SlideShare uma empresa Scribd logo
1 de 29
Baixar para ler offline
Generative Models as Distributions
of Functions
PR12와 함께 이해하는
Jaejun Yoo
(current) Postdoc. @EPFL
(from July) Assistant Prof., @UNIST
PR-312, 11th April, 2021
Today’s contents
“For all datasets, we use an MLP with 3
hidden layers of size 128 … and an MLP
with 2 hidden layers of size 256 and 512”
“We performed all training on a single
2080Ti GPU with 11GB of RAM.”
Motivation and Main Problem
“Conventional signal representations are usually discrete.”
However, Mother Nature is continuous!
(well… up to planck constant…?)
2D Images Audio 3D Shapes
Motivation and Main Problem
Of course, these functions are usually not analytically tractable. it is impossible to "write down"
the function that parameterizes a natural image as a mathematical formula.
Continuous representation?
Why hard?
Motivation and Main Problem
Why important?
• independent of spatial resolution (infinite resolution)
• Geometric transformation of images: zoom, rotation, super-resolution.
• Derivatives are well-defined.
Motivation and Main Problem
Why important?
• independent of spatial resolution (infinite resolution)
• Geometric transformation of images: zoom, rotation, super-resolution.
• Derivatives are well-defined.
Motivation and Main Problem
Why important?
Piecewise Constant Bilinear Cubic Spline
Motivation and Main Problem
Why important?
Piecewise constant Bilinear Cubic Spline
Motivation and Main Problem
Why important?
Motivation and Main Problem
Spline: Variational properties
Explicit derivative!
Motivation and Main Problem
Spline: Variational properties
Continuous representation?
• DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation (Park et al. 2019)
• Occupancy Networks: Learning 3D Reconstruction in Function Space (Mescheder et al. 2019)
• IM-Net: Learning Implicit Fields for Generative Shape Modeling (Chen et al. 2018)
• … NeRF (PR-302)…
“Implicit Neural Representations approximate this function via a neural network!”
Motivation and Main Problem
Implicit Neural Representation!
Implicit Neural Representation
- Remarkably, the representation !" is independent of
the number of pixels. The representation !" therefore,
unlike most image representations, does not depend
on the resolution of the image.
- The core property of these representations is that
they scale with signal complexity and not with
signal resolution.
Learning Distributions of Functions
1. Parameterizing a distribution over neural
networks with a hypernetwork (Ha et al., 2017)
Overall Scheme
“Sample the weights of a neural network”
to obtain a function.
: Learning a distribution over functions !" is equivalent to
learning a distribution over weights #(%).
: Then, #(%), where % = () * , is refer to as a neural
function distribution (NFD).
Learning Distributions of Functions
1. Parameterizing a distribution over neural
networks with a hypernetwork (Ha et al., 2017)
Overall Scheme
“Sample the weights of a neural network”
to obtain a function.
: Learning a distribution over functions !" is equivalent to
learning a distribution over weights #(%).
: Then, #(%), where % = () * , is refer to as a neural
function distribution (NFD).
However! How do we get access to the
ground truth functions to train the network?
Learning Distributions of Functions
1. Parameterizing a distribution over neural
networks with a hypernetwork (Ha et al., 2017)
Overall Scheme
“We do have access to input/output
pairs of these functions through the
coordinates and features, allowing us to
learn function distributions without
operating directly on the functions!”
Learning Distributions of Functions
1. Parameterizing a distribution over neural
networks with a hypernetwork (Ha et al., 2017)
Overall Scheme
2. Training this distribution with an adversarial
approach (Goodfellow et al., 2014).
“We do have access to input/output
pairs of these functions through the
coordinates and features, allowing us to
learn function distributions without
operating directly on the functions!”
Learning Distributions of Functions
1. Parameterizing a distribution over neural
networks with a hypernetwork (Ha et al., 2017)
Overall Scheme
2. Training this distribution with an adversarial
approach (Goodfellow et al., 2014).
* is a kind of position encoding (Fourier feature).
“We do have access to input/output
pairs of these functions through the
coordinates and features, allowing us to
learn function distributions without
operating directly on the functions!”
Learning Distributions of Functions
Overall Scheme
NFD
Now we know how to design a network to learn continuous functions!
Learning Distributions of Functions
Overall Scheme
Discriminator
But, the data we consider may not necessarily lie on a grid…
Learning Distributions of Functions
Overall Scheme
Discriminator
… in which case it is not possible to use convolutional discriminators.
Learning Distributions of Functions
Overall Scheme
Discriminator
Our discriminator should be able to distinguish between
real and fake sets of coordinate and feature pairs.
Point Cloud Discriminator
Point Convolution
In contrast to regular convolutions,
where the convolution kernels are only
defined at certain grid locations, the
convolution filters in PointConv are
parameterized by an MLP mapping
coordinates to kernel values:
Experiments
“For all datasets, we use an MLP with 3
hidden layers of size 128 … and an MLP
with 2 hidden layers of size 256 and 512”
“We performed all training on a single
2080Ti GPU with 11GB of RAM.”
“Remarkably, such a simple architecture
is sufficient for learning rich distributions
of images and 3D shapes.“
“Use the exact same model for both
images and 3D shapes except for the
input and output dimensions of the
function representation.”
Implementation Setups
Results
2D Image generation
• Samples from our model trained on CelebAHQ.
• 64×64 (top) and 128×128 (bottom)
• Each image corresponds to a function which
was sampled from our model and then
evaluated on the grid.
• To produce this figure we sampled 5 batches
and chose the best batch by visual inspection.
Results
“To the infinity and beyond!”
- Buzz Lightyear, Toy Story
Super-resolution
NFD
64×64
NFD
256×256
Bicubic
256×256
NFD
28×28
NFD
256×256
Bicubic
256×256
Results
3D shapes
Voxel grids from Choy et al. (2016) representing the chairs category from the ShapeNet (Chang et al.,
2015) dataset. The dataset contains 6778 chairs each of dimension 32#
. For each 3D model, uniformly
subsample K = 4096 points among 32# = 32,768 points and use them for training.
• A step towards making implicit neural representation methods genuinely useful
for modeling datasets rather than individual data points.
• The first framework to model data of this complexity in an entirely continuous
fashion.
• The ability of being independent to resolution and operating outside of a grid.
• A unique way of using point cloud discriminators.
Conclusion
Summary of Contributions (I think)
Things to discuss about…
• What kinds of study would be derived from this?
• Architectural developments (better quality)?
• Then How? Or what would be helpful?
• Other applications?
• Again, compute-driven AI vs human-knowledge based?
• Big model vs inductive bias?
• Etc.?

Mais conteúdo relacionado

Mais procurados

Diffusion models beat gans on image synthesis
Diffusion models beat gans on image synthesisDiffusion models beat gans on image synthesis
Diffusion models beat gans on image synthesisBeerenSahu
 
Tutorial on Deep Generative Models
 Tutorial on Deep Generative Models Tutorial on Deep Generative Models
Tutorial on Deep Generative ModelsMLReview
 
Deep Learning: Recurrent Neural Network (Chapter 10)
Deep Learning: Recurrent Neural Network (Chapter 10) Deep Learning: Recurrent Neural Network (Chapter 10)
Deep Learning: Recurrent Neural Network (Chapter 10) Larry Guo
 
Stable Diffusion path
Stable Diffusion pathStable Diffusion path
Stable Diffusion pathVitaly Bondar
 
Generative Adversarial Networks (GANs) - Ian Goodfellow, OpenAI
Generative Adversarial Networks (GANs) - Ian Goodfellow, OpenAIGenerative Adversarial Networks (GANs) - Ian Goodfellow, OpenAI
Generative Adversarial Networks (GANs) - Ian Goodfellow, OpenAIWithTheBest
 
Recurrent Neural Networks, LSTM and GRU
Recurrent Neural Networks, LSTM and GRURecurrent Neural Networks, LSTM and GRU
Recurrent Neural Networks, LSTM and GRUananth
 
Generative Adversarial Networks GAN - Xavier Giro - UPC TelecomBCN Barcelona ...
Generative Adversarial Networks GAN - Xavier Giro - UPC TelecomBCN Barcelona ...Generative Adversarial Networks GAN - Xavier Giro - UPC TelecomBCN Barcelona ...
Generative Adversarial Networks GAN - Xavier Giro - UPC TelecomBCN Barcelona ...Universitat Politècnica de Catalunya
 
Training Neural Networks
Training Neural NetworksTraining Neural Networks
Training Neural NetworksDatabricks
 
Feedforward neural network
Feedforward neural networkFeedforward neural network
Feedforward neural networkSopheaktra YONG
 
Recurrent Neural Networks (RNN) | RNN LSTM | Deep Learning Tutorial | Tensorf...
Recurrent Neural Networks (RNN) | RNN LSTM | Deep Learning Tutorial | Tensorf...Recurrent Neural Networks (RNN) | RNN LSTM | Deep Learning Tutorial | Tensorf...
Recurrent Neural Networks (RNN) | RNN LSTM | Deep Learning Tutorial | Tensorf...Edureka!
 
A brief introduction of Artificial neural network by example
A brief introduction of Artificial neural network by exampleA brief introduction of Artificial neural network by example
A brief introduction of Artificial neural network by exampleMrinmoy Majumder
 
“An Introduction to Data Augmentation Techniques in ML Frameworks,” a Present...
“An Introduction to Data Augmentation Techniques in ML Frameworks,” a Present...“An Introduction to Data Augmentation Techniques in ML Frameworks,” a Present...
“An Introduction to Data Augmentation Techniques in ML Frameworks,” a Present...Edge AI and Vision Alliance
 
ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]Dongmin Choi
 
Model-Based Reinforcement Learning @NIPS2017
Model-Based Reinforcement Learning @NIPS2017Model-Based Reinforcement Learning @NIPS2017
Model-Based Reinforcement Learning @NIPS2017mooopan
 
Master's Thesis Presentation
Master's Thesis PresentationMaster's Thesis Presentation
Master's Thesis PresentationWajdi Khattel
 
Introduction to Diffusion Models
Introduction to Diffusion ModelsIntroduction to Diffusion Models
Introduction to Diffusion ModelsSangwoo Mo
 
Introduction to MAML (Model Agnostic Meta Learning) with Discussions
Introduction to MAML (Model Agnostic Meta Learning) with DiscussionsIntroduction to MAML (Model Agnostic Meta Learning) with Discussions
Introduction to MAML (Model Agnostic Meta Learning) with DiscussionsJoonyoung Yi
 
Hyperparameter Tuning
Hyperparameter TuningHyperparameter Tuning
Hyperparameter TuningJon Lederman
 

Mais procurados (20)

Diffusion models beat gans on image synthesis
Diffusion models beat gans on image synthesisDiffusion models beat gans on image synthesis
Diffusion models beat gans on image synthesis
 
PhD Defense
PhD DefensePhD Defense
PhD Defense
 
Tutorial on Deep Generative Models
 Tutorial on Deep Generative Models Tutorial on Deep Generative Models
Tutorial on Deep Generative Models
 
Deep Learning: Recurrent Neural Network (Chapter 10)
Deep Learning: Recurrent Neural Network (Chapter 10) Deep Learning: Recurrent Neural Network (Chapter 10)
Deep Learning: Recurrent Neural Network (Chapter 10)
 
Stable Diffusion path
Stable Diffusion pathStable Diffusion path
Stable Diffusion path
 
Generative Adversarial Networks (GANs) - Ian Goodfellow, OpenAI
Generative Adversarial Networks (GANs) - Ian Goodfellow, OpenAIGenerative Adversarial Networks (GANs) - Ian Goodfellow, OpenAI
Generative Adversarial Networks (GANs) - Ian Goodfellow, OpenAI
 
Recurrent Neural Networks, LSTM and GRU
Recurrent Neural Networks, LSTM and GRURecurrent Neural Networks, LSTM and GRU
Recurrent Neural Networks, LSTM and GRU
 
Introduction to Transformer Model
Introduction to Transformer ModelIntroduction to Transformer Model
Introduction to Transformer Model
 
Generative Adversarial Networks GAN - Xavier Giro - UPC TelecomBCN Barcelona ...
Generative Adversarial Networks GAN - Xavier Giro - UPC TelecomBCN Barcelona ...Generative Adversarial Networks GAN - Xavier Giro - UPC TelecomBCN Barcelona ...
Generative Adversarial Networks GAN - Xavier Giro - UPC TelecomBCN Barcelona ...
 
Training Neural Networks
Training Neural NetworksTraining Neural Networks
Training Neural Networks
 
Feedforward neural network
Feedforward neural networkFeedforward neural network
Feedforward neural network
 
Recurrent Neural Networks (RNN) | RNN LSTM | Deep Learning Tutorial | Tensorf...
Recurrent Neural Networks (RNN) | RNN LSTM | Deep Learning Tutorial | Tensorf...Recurrent Neural Networks (RNN) | RNN LSTM | Deep Learning Tutorial | Tensorf...
Recurrent Neural Networks (RNN) | RNN LSTM | Deep Learning Tutorial | Tensorf...
 
A brief introduction of Artificial neural network by example
A brief introduction of Artificial neural network by exampleA brief introduction of Artificial neural network by example
A brief introduction of Artificial neural network by example
 
“An Introduction to Data Augmentation Techniques in ML Frameworks,” a Present...
“An Introduction to Data Augmentation Techniques in ML Frameworks,” a Present...“An Introduction to Data Augmentation Techniques in ML Frameworks,” a Present...
“An Introduction to Data Augmentation Techniques in ML Frameworks,” a Present...
 
ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]ViT (Vision Transformer) Review [CDM]
ViT (Vision Transformer) Review [CDM]
 
Model-Based Reinforcement Learning @NIPS2017
Model-Based Reinforcement Learning @NIPS2017Model-Based Reinforcement Learning @NIPS2017
Model-Based Reinforcement Learning @NIPS2017
 
Master's Thesis Presentation
Master's Thesis PresentationMaster's Thesis Presentation
Master's Thesis Presentation
 
Introduction to Diffusion Models
Introduction to Diffusion ModelsIntroduction to Diffusion Models
Introduction to Diffusion Models
 
Introduction to MAML (Model Agnostic Meta Learning) with Discussions
Introduction to MAML (Model Agnostic Meta Learning) with DiscussionsIntroduction to MAML (Model Agnostic Meta Learning) with Discussions
Introduction to MAML (Model Agnostic Meta Learning) with Discussions
 
Hyperparameter Tuning
Hyperparameter TuningHyperparameter Tuning
Hyperparameter Tuning
 

Semelhante a [PR12] Generative Models as Distributions of Functions

Machine learning in science and industry — day 4
Machine learning in science and industry — day 4Machine learning in science and industry — day 4
Machine learning in science and industry — day 4arogozhnikov
 
Automatic Attendace using convolutional neural network Face Recognition
Automatic Attendace using convolutional neural network Face RecognitionAutomatic Attendace using convolutional neural network Face Recognition
Automatic Attendace using convolutional neural network Face Recognitionvatsal199567
 
Variants of GANs - Jaejun Yoo
Variants of GANs - Jaejun YooVariants of GANs - Jaejun Yoo
Variants of GANs - Jaejun YooJaeJun Yoo
 
Crafting Recommenders: the Shallow and the Deep of it!
Crafting Recommenders: the Shallow and the Deep of it! Crafting Recommenders: the Shallow and the Deep of it!
Crafting Recommenders: the Shallow and the Deep of it! Sudeep Das, Ph.D.
 
Disentangled Representation Learning of Deep Generative Models
Disentangled Representation Learning of Deep Generative ModelsDisentangled Representation Learning of Deep Generative Models
Disentangled Representation Learning of Deep Generative ModelsRyohei Suzuki
 
PR095: Modularity Matters: Learning Invariant Relational Reasoning Tasks
PR095: Modularity Matters: Learning Invariant Relational Reasoning TasksPR095: Modularity Matters: Learning Invariant Relational Reasoning Tasks
PR095: Modularity Matters: Learning Invariant Relational Reasoning TasksJinwon Lee
 
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 GANsParham Zilouchian
 
[PR12] understanding deep learning requires rethinking generalization
[PR12] understanding deep learning requires rethinking generalization[PR12] understanding deep learning requires rethinking generalization
[PR12] understanding deep learning requires rethinking generalizationJaeJun Yoo
 
introduction to deeplearning
introduction to deeplearningintroduction to deeplearning
introduction to deeplearningEyad Alshami
 
Deep Learning and Reinforcement Learning
Deep Learning and Reinforcement LearningDeep Learning and Reinforcement Learning
Deep Learning and Reinforcement LearningRenārs Liepiņš
 
Image classification with Deep Neural Networks
Image classification with Deep Neural NetworksImage classification with Deep Neural Networks
Image classification with Deep Neural NetworksYogendra Tamang
 
The Search for a New Visual Search Beyond Language - StampedeCon AI Summit 2017
The Search for a New Visual Search Beyond Language - StampedeCon AI Summit 2017The Search for a New Visual Search Beyond Language - StampedeCon AI Summit 2017
The Search for a New Visual Search Beyond Language - StampedeCon AI Summit 2017StampedeCon
 
Action Genome: Action As Composition of Spatio Temporal Scene Graphs
Action Genome: Action As Composition of Spatio Temporal Scene GraphsAction Genome: Action As Composition of Spatio Temporal Scene Graphs
Action Genome: Action As Composition of Spatio Temporal Scene GraphsSangmin Woo
 
NIPS2007: deep belief nets
NIPS2007: deep belief netsNIPS2007: deep belief nets
NIPS2007: deep belief netszukun
 
MLIP - Chapter 3 - Introduction to deep learning
MLIP - Chapter 3 - Introduction to deep learningMLIP - Chapter 3 - Introduction to deep learning
MLIP - Chapter 3 - Introduction to deep learningCharles Deledalle
 
Deep learning tutorial 9/2019
Deep learning tutorial 9/2019Deep learning tutorial 9/2019
Deep learning tutorial 9/2019Amr Rashed
 
Deep Learning Tutorial
Deep Learning TutorialDeep Learning Tutorial
Deep Learning TutorialAmr Rashed
 
JPM1410 Images as Occlusions of Textures: A Framework for Segmentation
JPM1410   Images as Occlusions of Textures: A Framework for SegmentationJPM1410   Images as Occlusions of Textures: A Framework for Segmentation
JPM1410 Images as Occlusions of Textures: A Framework for Segmentationchennaijp
 
Evolution of Deep Learning and new advancements
Evolution of Deep Learning and new advancementsEvolution of Deep Learning and new advancements
Evolution of Deep Learning and new advancementsChitta Ranjan
 

Semelhante a [PR12] Generative Models as Distributions of Functions (20)

Image captioning
Image captioningImage captioning
Image captioning
 
Machine learning in science and industry — day 4
Machine learning in science and industry — day 4Machine learning in science and industry — day 4
Machine learning in science and industry — day 4
 
Automatic Attendace using convolutional neural network Face Recognition
Automatic Attendace using convolutional neural network Face RecognitionAutomatic Attendace using convolutional neural network Face Recognition
Automatic Attendace using convolutional neural network Face Recognition
 
Variants of GANs - Jaejun Yoo
Variants of GANs - Jaejun YooVariants of GANs - Jaejun Yoo
Variants of GANs - Jaejun Yoo
 
Crafting Recommenders: the Shallow and the Deep of it!
Crafting Recommenders: the Shallow and the Deep of it! Crafting Recommenders: the Shallow and the Deep of it!
Crafting Recommenders: the Shallow and the Deep of it!
 
Disentangled Representation Learning of Deep Generative Models
Disentangled Representation Learning of Deep Generative ModelsDisentangled Representation Learning of Deep Generative Models
Disentangled Representation Learning of Deep Generative Models
 
PR095: Modularity Matters: Learning Invariant Relational Reasoning Tasks
PR095: Modularity Matters: Learning Invariant Relational Reasoning TasksPR095: Modularity Matters: Learning Invariant Relational Reasoning Tasks
PR095: Modularity Matters: Learning Invariant Relational Reasoning Tasks
 
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
 
[PR12] understanding deep learning requires rethinking generalization
[PR12] understanding deep learning requires rethinking generalization[PR12] understanding deep learning requires rethinking generalization
[PR12] understanding deep learning requires rethinking generalization
 
introduction to deeplearning
introduction to deeplearningintroduction to deeplearning
introduction to deeplearning
 
Deep Learning and Reinforcement Learning
Deep Learning and Reinforcement LearningDeep Learning and Reinforcement Learning
Deep Learning and Reinforcement Learning
 
Image classification with Deep Neural Networks
Image classification with Deep Neural NetworksImage classification with Deep Neural Networks
Image classification with Deep Neural Networks
 
The Search for a New Visual Search Beyond Language - StampedeCon AI Summit 2017
The Search for a New Visual Search Beyond Language - StampedeCon AI Summit 2017The Search for a New Visual Search Beyond Language - StampedeCon AI Summit 2017
The Search for a New Visual Search Beyond Language - StampedeCon AI Summit 2017
 
Action Genome: Action As Composition of Spatio Temporal Scene Graphs
Action Genome: Action As Composition of Spatio Temporal Scene GraphsAction Genome: Action As Composition of Spatio Temporal Scene Graphs
Action Genome: Action As Composition of Spatio Temporal Scene Graphs
 
NIPS2007: deep belief nets
NIPS2007: deep belief netsNIPS2007: deep belief nets
NIPS2007: deep belief nets
 
MLIP - Chapter 3 - Introduction to deep learning
MLIP - Chapter 3 - Introduction to deep learningMLIP - Chapter 3 - Introduction to deep learning
MLIP - Chapter 3 - Introduction to deep learning
 
Deep learning tutorial 9/2019
Deep learning tutorial 9/2019Deep learning tutorial 9/2019
Deep learning tutorial 9/2019
 
Deep Learning Tutorial
Deep Learning TutorialDeep Learning Tutorial
Deep Learning Tutorial
 
JPM1410 Images as Occlusions of Textures: A Framework for Segmentation
JPM1410   Images as Occlusions of Textures: A Framework for SegmentationJPM1410   Images as Occlusions of Textures: A Framework for Segmentation
JPM1410 Images as Occlusions of Textures: A Framework for Segmentation
 
Evolution of Deep Learning and new advancements
Evolution of Deep Learning and new advancementsEvolution of Deep Learning and new advancements
Evolution of Deep Learning and new advancements
 

Mais de JaeJun Yoo

[CVPR2020] Simple but effective image enhancement techniques
[CVPR2020] Simple but effective image enhancement techniques[CVPR2020] Simple but effective image enhancement techniques
[CVPR2020] Simple but effective image enhancement techniquesJaeJun Yoo
 
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...JaeJun Yoo
 
Super resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun YooSuper resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun YooJaeJun Yoo
 
A beginner's guide to Style Transfer and recent trends
A beginner's guide to Style Transfer and recent trendsA beginner's guide to Style Transfer and recent trends
A beginner's guide to Style Transfer and recent trendsJaeJun Yoo
 
[PR12] Spectral Normalization for Generative Adversarial Networks
[PR12] Spectral Normalization for Generative Adversarial Networks[PR12] Spectral Normalization for Generative Adversarial Networks
[PR12] Spectral Normalization for Generative Adversarial NetworksJaeJun Yoo
 
Introduction to ambient GAN
Introduction to ambient GANIntroduction to ambient GAN
Introduction to ambient GANJaeJun Yoo
 
[PR12] categorical reparameterization with gumbel softmax
[PR12] categorical reparameterization with gumbel softmax[PR12] categorical reparameterization with gumbel softmax
[PR12] categorical reparameterization with gumbel softmaxJaeJun Yoo
 
[PR12] Capsule Networks - Jaejun Yoo
[PR12] Capsule Networks - Jaejun Yoo[PR12] Capsule Networks - Jaejun Yoo
[PR12] Capsule Networks - Jaejun YooJaeJun Yoo
 
[PR12] Inception and Xception - Jaejun Yoo
[PR12] Inception and Xception - Jaejun Yoo[PR12] Inception and Xception - Jaejun Yoo
[PR12] Inception and Xception - Jaejun YooJaeJun Yoo
 
[PR12] PixelRNN- Jaejun Yoo
[PR12] PixelRNN- Jaejun Yoo[PR12] PixelRNN- Jaejun Yoo
[PR12] PixelRNN- Jaejun YooJaeJun Yoo
 
[Pr12] dann jaejun yoo
[Pr12] dann   jaejun yoo[Pr12] dann   jaejun yoo
[Pr12] dann jaejun yooJaeJun Yoo
 
[PR12] intro. to gans jaejun yoo
[PR12] intro. to gans   jaejun yoo[PR12] intro. to gans   jaejun yoo
[PR12] intro. to gans jaejun yooJaeJun Yoo
 

Mais de JaeJun Yoo (12)

[CVPR2020] Simple but effective image enhancement techniques
[CVPR2020] Simple but effective image enhancement techniques[CVPR2020] Simple but effective image enhancement techniques
[CVPR2020] Simple but effective image enhancement techniques
 
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...
Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Anal...
 
Super resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun YooSuper resolution in deep learning era - Jaejun Yoo
Super resolution in deep learning era - Jaejun Yoo
 
A beginner's guide to Style Transfer and recent trends
A beginner's guide to Style Transfer and recent trendsA beginner's guide to Style Transfer and recent trends
A beginner's guide to Style Transfer and recent trends
 
[PR12] Spectral Normalization for Generative Adversarial Networks
[PR12] Spectral Normalization for Generative Adversarial Networks[PR12] Spectral Normalization for Generative Adversarial Networks
[PR12] Spectral Normalization for Generative Adversarial Networks
 
Introduction to ambient GAN
Introduction to ambient GANIntroduction to ambient GAN
Introduction to ambient GAN
 
[PR12] categorical reparameterization with gumbel softmax
[PR12] categorical reparameterization with gumbel softmax[PR12] categorical reparameterization with gumbel softmax
[PR12] categorical reparameterization with gumbel softmax
 
[PR12] Capsule Networks - Jaejun Yoo
[PR12] Capsule Networks - Jaejun Yoo[PR12] Capsule Networks - Jaejun Yoo
[PR12] Capsule Networks - Jaejun Yoo
 
[PR12] Inception and Xception - Jaejun Yoo
[PR12] Inception and Xception - Jaejun Yoo[PR12] Inception and Xception - Jaejun Yoo
[PR12] Inception and Xception - Jaejun Yoo
 
[PR12] PixelRNN- Jaejun Yoo
[PR12] PixelRNN- Jaejun Yoo[PR12] PixelRNN- Jaejun Yoo
[PR12] PixelRNN- Jaejun Yoo
 
[Pr12] dann jaejun yoo
[Pr12] dann   jaejun yoo[Pr12] dann   jaejun yoo
[Pr12] dann jaejun yoo
 
[PR12] intro. to gans jaejun yoo
[PR12] intro. to gans   jaejun yoo[PR12] intro. to gans   jaejun yoo
[PR12] intro. to gans jaejun yoo
 

Último

POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.Silpa
 
An introduction on sequence tagged site mapping
An introduction on sequence tagged site mappingAn introduction on sequence tagged site mapping
An introduction on sequence tagged site mappingadibshanto115
 
GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)Areesha Ahmad
 
Conjugation, transduction and transformation
Conjugation, transduction and transformationConjugation, transduction and transformation
Conjugation, transduction and transformationAreesha Ahmad
 
pumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit flypumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit flyPRADYUMMAURYA1
 
Call Girls Ahmedabad +917728919243 call me Independent Escort Service
Call Girls Ahmedabad +917728919243 call me Independent Escort ServiceCall Girls Ahmedabad +917728919243 call me Independent Escort Service
Call Girls Ahmedabad +917728919243 call me Independent Escort Serviceshivanisharma5244
 
Bacterial Identification and Classifications
Bacterial Identification and ClassificationsBacterial Identification and Classifications
Bacterial Identification and ClassificationsAreesha Ahmad
 
Molecular markers- RFLP, RAPD, AFLP, SNP etc.
Molecular markers- RFLP, RAPD, AFLP, SNP etc.Molecular markers- RFLP, RAPD, AFLP, SNP etc.
Molecular markers- RFLP, RAPD, AFLP, SNP etc.Silpa
 
Thyroid Physiology_Dr.E. Muralinath_ Associate Professor
Thyroid Physiology_Dr.E. Muralinath_ Associate ProfessorThyroid Physiology_Dr.E. Muralinath_ Associate Professor
Thyroid Physiology_Dr.E. Muralinath_ Associate Professormuralinath2
 
300003-World Science Day For Peace And Development.pptx
300003-World Science Day For Peace And Development.pptx300003-World Science Day For Peace And Development.pptx
300003-World Science Day For Peace And Development.pptxryanrooker
 
PSYCHOSOCIAL NEEDS. in nursing II sem pptx
PSYCHOSOCIAL NEEDS. in nursing II sem pptxPSYCHOSOCIAL NEEDS. in nursing II sem pptx
PSYCHOSOCIAL NEEDS. in nursing II sem pptxSuji236384
 
Zoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdfZoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdfSumit Kumar yadav
 
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...Silpa
 
Exploring Criminology and Criminal Behaviour.pdf
Exploring Criminology and Criminal Behaviour.pdfExploring Criminology and Criminal Behaviour.pdf
Exploring Criminology and Criminal Behaviour.pdfrohankumarsinghrore1
 
development of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virusdevelopment of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virusNazaninKarimi6
 
FAIRSpectra - Enabling the FAIRification of Analytical Science
FAIRSpectra - Enabling the FAIRification of Analytical ScienceFAIRSpectra - Enabling the FAIRification of Analytical Science
FAIRSpectra - Enabling the FAIRification of Analytical ScienceAlex Henderson
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsSérgio Sacani
 
Grade 7 - Lesson 1 - Microscope and Its Functions
Grade 7 - Lesson 1 - Microscope and Its FunctionsGrade 7 - Lesson 1 - Microscope and Its Functions
Grade 7 - Lesson 1 - Microscope and Its FunctionsOrtegaSyrineMay
 

Último (20)

POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.
 
An introduction on sequence tagged site mapping
An introduction on sequence tagged site mappingAn introduction on sequence tagged site mapping
An introduction on sequence tagged site mapping
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)
 
Conjugation, transduction and transformation
Conjugation, transduction and transformationConjugation, transduction and transformation
Conjugation, transduction and transformation
 
pumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit flypumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit fly
 
Call Girls Ahmedabad +917728919243 call me Independent Escort Service
Call Girls Ahmedabad +917728919243 call me Independent Escort ServiceCall Girls Ahmedabad +917728919243 call me Independent Escort Service
Call Girls Ahmedabad +917728919243 call me Independent Escort Service
 
Site Acceptance Test .
Site Acceptance Test                    .Site Acceptance Test                    .
Site Acceptance Test .
 
Bacterial Identification and Classifications
Bacterial Identification and ClassificationsBacterial Identification and Classifications
Bacterial Identification and Classifications
 
Molecular markers- RFLP, RAPD, AFLP, SNP etc.
Molecular markers- RFLP, RAPD, AFLP, SNP etc.Molecular markers- RFLP, RAPD, AFLP, SNP etc.
Molecular markers- RFLP, RAPD, AFLP, SNP etc.
 
Thyroid Physiology_Dr.E. Muralinath_ Associate Professor
Thyroid Physiology_Dr.E. Muralinath_ Associate ProfessorThyroid Physiology_Dr.E. Muralinath_ Associate Professor
Thyroid Physiology_Dr.E. Muralinath_ Associate Professor
 
300003-World Science Day For Peace And Development.pptx
300003-World Science Day For Peace And Development.pptx300003-World Science Day For Peace And Development.pptx
300003-World Science Day For Peace And Development.pptx
 
PSYCHOSOCIAL NEEDS. in nursing II sem pptx
PSYCHOSOCIAL NEEDS. in nursing II sem pptxPSYCHOSOCIAL NEEDS. in nursing II sem pptx
PSYCHOSOCIAL NEEDS. in nursing II sem pptx
 
Zoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdfZoology 5th semester notes( Sumit_yadav).pdf
Zoology 5th semester notes( Sumit_yadav).pdf
 
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...
Locating and isolating a gene, FISH, GISH, Chromosome walking and jumping, te...
 
Exploring Criminology and Criminal Behaviour.pdf
Exploring Criminology and Criminal Behaviour.pdfExploring Criminology and Criminal Behaviour.pdf
Exploring Criminology and Criminal Behaviour.pdf
 
development of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virusdevelopment of diagnostic enzyme assay to detect leuser virus
development of diagnostic enzyme assay to detect leuser virus
 
FAIRSpectra - Enabling the FAIRification of Analytical Science
FAIRSpectra - Enabling the FAIRification of Analytical ScienceFAIRSpectra - Enabling the FAIRification of Analytical Science
FAIRSpectra - Enabling the FAIRification of Analytical Science
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
 
Grade 7 - Lesson 1 - Microscope and Its Functions
Grade 7 - Lesson 1 - Microscope and Its FunctionsGrade 7 - Lesson 1 - Microscope and Its Functions
Grade 7 - Lesson 1 - Microscope and Its Functions
 

[PR12] Generative Models as Distributions of Functions

  • 1. Generative Models as Distributions of Functions PR12와 함께 이해하는 Jaejun Yoo (current) Postdoc. @EPFL (from July) Assistant Prof., @UNIST PR-312, 11th April, 2021
  • 2. Today’s contents “For all datasets, we use an MLP with 3 hidden layers of size 128 … and an MLP with 2 hidden layers of size 256 and 512” “We performed all training on a single 2080Ti GPU with 11GB of RAM.”
  • 3. Motivation and Main Problem “Conventional signal representations are usually discrete.” However, Mother Nature is continuous! (well… up to planck constant…?) 2D Images Audio 3D Shapes
  • 4. Motivation and Main Problem Of course, these functions are usually not analytically tractable. it is impossible to "write down" the function that parameterizes a natural image as a mathematical formula. Continuous representation? Why hard?
  • 5. Motivation and Main Problem Why important? • independent of spatial resolution (infinite resolution) • Geometric transformation of images: zoom, rotation, super-resolution. • Derivatives are well-defined.
  • 6. Motivation and Main Problem Why important? • independent of spatial resolution (infinite resolution) • Geometric transformation of images: zoom, rotation, super-resolution. • Derivatives are well-defined.
  • 7. Motivation and Main Problem Why important? Piecewise Constant Bilinear Cubic Spline
  • 8. Motivation and Main Problem Why important? Piecewise constant Bilinear Cubic Spline
  • 9. Motivation and Main Problem Why important?
  • 10. Motivation and Main Problem Spline: Variational properties Explicit derivative!
  • 11. Motivation and Main Problem Spline: Variational properties
  • 12. Continuous representation? • DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation (Park et al. 2019) • Occupancy Networks: Learning 3D Reconstruction in Function Space (Mescheder et al. 2019) • IM-Net: Learning Implicit Fields for Generative Shape Modeling (Chen et al. 2018) • … NeRF (PR-302)… “Implicit Neural Representations approximate this function via a neural network!” Motivation and Main Problem Implicit Neural Representation!
  • 13. Implicit Neural Representation - Remarkably, the representation !" is independent of the number of pixels. The representation !" therefore, unlike most image representations, does not depend on the resolution of the image. - The core property of these representations is that they scale with signal complexity and not with signal resolution.
  • 14. Learning Distributions of Functions 1. Parameterizing a distribution over neural networks with a hypernetwork (Ha et al., 2017) Overall Scheme “Sample the weights of a neural network” to obtain a function. : Learning a distribution over functions !" is equivalent to learning a distribution over weights #(%). : Then, #(%), where % = () * , is refer to as a neural function distribution (NFD).
  • 15. Learning Distributions of Functions 1. Parameterizing a distribution over neural networks with a hypernetwork (Ha et al., 2017) Overall Scheme “Sample the weights of a neural network” to obtain a function. : Learning a distribution over functions !" is equivalent to learning a distribution over weights #(%). : Then, #(%), where % = () * , is refer to as a neural function distribution (NFD). However! How do we get access to the ground truth functions to train the network?
  • 16. Learning Distributions of Functions 1. Parameterizing a distribution over neural networks with a hypernetwork (Ha et al., 2017) Overall Scheme “We do have access to input/output pairs of these functions through the coordinates and features, allowing us to learn function distributions without operating directly on the functions!”
  • 17. Learning Distributions of Functions 1. Parameterizing a distribution over neural networks with a hypernetwork (Ha et al., 2017) Overall Scheme 2. Training this distribution with an adversarial approach (Goodfellow et al., 2014). “We do have access to input/output pairs of these functions through the coordinates and features, allowing us to learn function distributions without operating directly on the functions!”
  • 18. Learning Distributions of Functions 1. Parameterizing a distribution over neural networks with a hypernetwork (Ha et al., 2017) Overall Scheme 2. Training this distribution with an adversarial approach (Goodfellow et al., 2014). * is a kind of position encoding (Fourier feature). “We do have access to input/output pairs of these functions through the coordinates and features, allowing us to learn function distributions without operating directly on the functions!”
  • 19. Learning Distributions of Functions Overall Scheme NFD Now we know how to design a network to learn continuous functions!
  • 20. Learning Distributions of Functions Overall Scheme Discriminator But, the data we consider may not necessarily lie on a grid…
  • 21. Learning Distributions of Functions Overall Scheme Discriminator … in which case it is not possible to use convolutional discriminators.
  • 22. Learning Distributions of Functions Overall Scheme Discriminator Our discriminator should be able to distinguish between real and fake sets of coordinate and feature pairs.
  • 23. Point Cloud Discriminator Point Convolution In contrast to regular convolutions, where the convolution kernels are only defined at certain grid locations, the convolution filters in PointConv are parameterized by an MLP mapping coordinates to kernel values:
  • 24. Experiments “For all datasets, we use an MLP with 3 hidden layers of size 128 … and an MLP with 2 hidden layers of size 256 and 512” “We performed all training on a single 2080Ti GPU with 11GB of RAM.” “Remarkably, such a simple architecture is sufficient for learning rich distributions of images and 3D shapes.“ “Use the exact same model for both images and 3D shapes except for the input and output dimensions of the function representation.” Implementation Setups
  • 25. Results 2D Image generation • Samples from our model trained on CelebAHQ. • 64×64 (top) and 128×128 (bottom) • Each image corresponds to a function which was sampled from our model and then evaluated on the grid. • To produce this figure we sampled 5 batches and chose the best batch by visual inspection.
  • 26. Results “To the infinity and beyond!” - Buzz Lightyear, Toy Story Super-resolution NFD 64×64 NFD 256×256 Bicubic 256×256 NFD 28×28 NFD 256×256 Bicubic 256×256
  • 27. Results 3D shapes Voxel grids from Choy et al. (2016) representing the chairs category from the ShapeNet (Chang et al., 2015) dataset. The dataset contains 6778 chairs each of dimension 32# . For each 3D model, uniformly subsample K = 4096 points among 32# = 32,768 points and use them for training.
  • 28. • A step towards making implicit neural representation methods genuinely useful for modeling datasets rather than individual data points. • The first framework to model data of this complexity in an entirely continuous fashion. • The ability of being independent to resolution and operating outside of a grid. • A unique way of using point cloud discriminators. Conclusion Summary of Contributions (I think)
  • 29. Things to discuss about… • What kinds of study would be derived from this? • Architectural developments (better quality)? • Then How? Or what would be helpful? • Other applications? • Again, compute-driven AI vs human-knowledge based? • Big model vs inductive bias? • Etc.?