PixelGAN Autoencoders

Slides:



Advertisements
Similar presentations
Part 2: Unsupervised Learning
Advertisements

Applications of one-class classification
Unsupervised Learning Clustering K-Means. Recall: Key Components of Intelligent Agents Representation Language: Graph, Bayes Nets, Linear functions Inference.
Salvatore giorgi Ece 8110 machine learning 5/12/2014
Pitch Prediction From MFCC Vectors for Speech Reconstruction Xu shao and Ben Milner School of Computing Sciences, University of East Anglia, UK Presented.
Automatic Identification of Bacterial Types using Statistical Image Modeling Sigal Trattner, Dr. Hayit Greenspan, Prof. Shimon Abboud Department of Biomedical.
Relational Learning with Gaussian Processes By Wei Chu, Vikas Sindhwani, Zoubin Ghahramani, S.Sathiya Keerthi (Columbia, Chicago, Cambridge, Yahoo!) Presented.
Joint Estimation of Image Clusters and Image Transformations Brendan J. Frey Computer Science, University of Waterloo, Canada Beckman Institute and ECE,
ETHEM ALPAYDIN © The MIT Press, Lecture Slides for.
Autoencoders Mostafa Heidarpour
Image Denoising and Inpainting with Deep Neural Networks Junyuan Xie, Linli Xu, Enhong Chen School of Computer Science and Technology University of Science.
INTRODUCTION TO Machine Learning ETHEM ALPAYDIN © The MIT Press, Lecture Slides for.
Introduction to machine learning
Latent Variable Models Christopher M. Bishop. 1. Density Modeling A standard approach: parametric models  a number of adaptive parameters  Gaussian.
Transfer Learning Task. Problem Identification Dataset : A Year: 2000 Features: 48 Training Model ‘M’ Testing 98.6% Training Model ‘M’ Testing 97% Dataset.
Jun-Won Suh Intelligent Electronic Systems Human and Systems Engineering Department of Electrical and Computer Engineering Speaker Verification System.
MACHINE LEARNING 8. Clustering. Motivation Based on E ALPAYDIN 2004 Introduction to Machine Learning © The MIT Press (V1.1) 2  Classification problem:
Students: Meera & Si Mentor: Afshin Dehghan WEEK 4: DEEP TRACKING.
Topic Models Presented by Iulian Pruteanu Friday, July 28 th, 2006.
Intelligent Database Systems Lab Advisor : Dr.Hsu Graduate : Keng-Wei Chang Author : Lian Yan and David J. Miller 國立雲林科技大學 National Yunlin University of.
CSC2515: Lecture 7 (post) Independent Components Analysis, and Autoencoders Geoffrey Hinton.
Lecture 2: Statistical learning primer for biologists
1 Unsupervised Learning and Clustering Shyh-Kang Jeng Department of Electrical Engineering/ Graduate Institute of Communication/ Graduate Institute of.
Face detection and recognition Many slides adapted from K. Grauman and D. Lowe.
Variational Autoencoders Presentation by Yuri Burda CS2523, University of Toronto.
Anomaly Detection in Data Science
Unsupervised Learning of Video Representations using LSTMs
WAVENET: A GENERATIVE MODEL FOR RAW AUDIO
Automatic Lung Cancer Diagnosis from CT Scans (Week 2)
Data Mining, Neural Network and Genetic Programming
Learning Deep L0 Encoders
Deep Predictive Model for Autonomous Driving
An Additive Latent Feature Model
Automatic Lung Cancer Diagnosis from CT Scans (Week 4)
Perceptual Loss Deep Feature Interpolation for Image Content Changes
CNN Demo LIU Pengpeng.
Neural networks (3) Regularization Autoencoder
Deep learning and applications to Natural language processing
Presenter: Hajar Emami
Overview of Supervised Learning
Unsupervised Learning and Autoencoders
Adversarially Tuned Scene Generation
Vision Reading Group, 2nd of May 2018 Martin Rünz
LTI Student Research Symposium 2004 Antoine Raux
Probabilistic Models with Latent Variables
(boy, that’s a long title)
ECE 599/692 – Deep Learning Lecture 9 – Autoencoder (AE)
Transformation-invariant clustering using the EM algorithm
Hairong Qi, Gonzalez Family Professor
Overview of Machine Learning
CS 2750: Machine Learning Expectation Maximization
Supervised vs. unsupervised Learning
Image to Image Translation using GANs
INTRODUCTION TO Machine Learning
View Inter-Prediction GAN: Unsupervised Representation Learning for 3D Shapes by Learning Global Shape Memories to Support Local View Predictions 1,2 1.
Ernest Valveny Computer Vision Center
Machine learning overview
Zhedong Zheng, Liang Zheng and Yi Yang
Using Manifold Structure for Partially Labeled Classification
Compressive Image Recovery using Recurrent Generative Model
Machine Learning – a Probabilistic Perspective
Report Yang Zhang.
one-input multi-output architecture
Hsiao-Yu Chiang Xiaoya Li Ganyu “Bruce” Xu Pinshuo Ye Zhanyuan Zhang
Background Task Fashion image inpainting Some conceptions
CSC 578 Neural Networks and Deep Learning
Goodfellow: Chapter 14 Autoencoders
Model based RL Part 1.
Presentation transcript:

PixelGAN Autoencoders Alireza Makhzani, Brendan Frey University of Toronto Liu ze Dec 30th, 2017 中国科学技术大学 University of Science and Technology of China

Outline 1. Background 2. PixelGAN Autoencoders 3. Experiments PixelCNNs Variational Autoencoders Adversarial Autoencoders 2. PixelGAN Autoencoders Limitations of VAE/AAE Structure and Training Benefits of PixelGAN Autoencoders 3. Experiments 4. Conclusion

Outline 1. Background 2. PixelGAN Autoencoders 3. Experiments PixelCNNs Variational Autoencoders Adversarial Autoencoders 2. PixelGAN Autoencoders Limitations of VAE/AAE Structure and Training Benefits of PixelGAN Autoencoders 3. Experiments 4. Conclusion

PixelCNNs

Contional PixelCNNs h h

Conditional PixelCNNs h h Learn the image statistics directly at the pixel level. Good at modelling low-level pixel statistics. Conditional PixelCNNs can learn conditional densities. Samples lack global structure. Lacking latent representation.

Variational Autoencoders Good at capturing the global structure, but samples are blurry.

Adversarial Autoencoders Code Space of MNIST: Gaussian Prior Mixture of Gaussians

Outline 1. Background 2. PixelGAN Autoencoders 3. Experiments PixelCNNs Variational Autoencoders Adversarial Autoencoders 2. PixelGAN Autoencoders Limitations of VAE/AAE Structure and Training Benefits of PixelGAN Autoencoders 3. Experiments 4. Conclusion

Limitations of VAE/AAE ✦ All the image statistics are captured by the single latent vector. VAE label, style global and local p(z) Latent Variable Deterministic (factorized Gaussians) p(x|z) None

Structure and Training Cost function of PixelGAN = Reconstruction + Adversarial Cost

Benefits of PixelGAN Autoencoders ✦ The image statistics are captured jointly by the latent vector and the autoregressive decoder. p(z) Latent Variable p(x|z) PixelCNN

Benefits of PixelGAN Autoencoders ✦ The image statistics are captured jointly by the latent vector and the autoregressive decoder. PixelGAN (Gaussian) PixelGAN (Categorical) Discrete (label) Global (low-frequency) p(z) Latent Variable Local (high-frequency) Continuous (Style) p(x|z) PixelCNN

PixelGAN (Categorical) Benefits of PixelGAN Autoencoders ✦ The image statistics are captured jointly by the latent vector and the autoregressive decoder. PixelGAN (Gaussian) PixelGAN (Categorical) Discrete (label) Global (low-frequency) p(z) Latent Variable Local (high-frequency) Continuous (Style) p(x|z) PixelCNN Semi-supervised Learning

Outline 1. Background 2. PixelGAN Autoencoders 3. Experiments PixelCNNs Variational Autoencoders Adversarial Autoencoders 2. PixelGAN Autoencoders Limitations of VAE/AAE Structure and Training Benefits of PixelGAN Autoencoders 3. Experiments 4. Conclusion

Global vs. Local Decomposition

Code Space Code Space of MNIST:

PixelGAN Autoencoders with Categorical Priors

Discrete vs. Continuous Decomposition (Clustering)

Discrete vs. Continuous Decomposition (Clustering)

Unsupervised Clustering 5% Error rate

Semi-supervised Learning

Semi-supervised Classification

Outline 1. Background 2. PixelGAN Autoencoders 3. Experiments PixelCNNs Variational Autoencoders Adversarial Autoencoders 2. PixelGAN Autoencoders Limitations of VAE/AAE Structure and Training Benefits of PixelGAN Autoencoders 3. Experiments 4. Conclusion

Unsupervised Clustering A Proposed the PixelGAN autoencoder, which is a generative autoencoder that combines a generative PixelCNN with a GAN inference network that can impose arbitrary priors on the latent code. B Showed that imposing different distributions as the prior enables us to learn a latent representation that captures the type of statistics that we care about, while the remaining structure of the image is captured by the PixelCNN decoder. C Demonstrate the application of PixelGAN autoencoders in downstream tasks such as semi-supervised learning; Discussed how these architectures have other potentials such as learning cross-domain relations between two different domains

Thank you!