Closed NorbertZheng closed 1 year ago
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks.
In this story, Deep Convolutional Generative Adversarial Network (DCGAN), by Indico Research and Facebook AI Research (FAIR), is reviewed. With DCGAN, a hierarchy of representations is learned from object parts to scenes in both the generator and discriminator. This is a paper in 2016 ICLR with about 6000 citations.
Three datasets are trained: LSUN, ImageNet, and a newly assembled faces dataset.
Generated bedrooms after one epoch. Generated bedrooms after five epochs of training.
A model is trained on the LSUN bedrooms dataset containing a little over 3 million training examples.
The model is not producing high quality samples via simply overfitting/memorizing training examples.
Accuracy (%) on CIFAR10.
DCGAN is trained on ImageNet-1k and then
These features are then flattened and concatenated to form a 28672 dimensional vector and a regularized linear L2-SVM classifier is trained on top of them.
Interpolation between a series of 9 random points in Z show that the space learned has smooth transitions.
Left: Random Filter Baseline, Right: Trained Filters.
First Row: Models Without Dropping “Window” Filters, Second Row: Models With Dropping “Window” Filters.
Smiling Woman — Neutral Woman + Neutral Man = Smiling Man.
Man With Glasses — Man Without Glasses + Woman Without Glasses = Woman With Glasses.
A ”turn” vector was created from four averaged samples of faces looking left vs looking right.
Vector Arithmetic on Input Space.
[2016 ICLR] [DCGAN] Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks.
Sik-Ho Tsang. Review: DCGAN — Deep Convolutional Generative Adversarial Network (GAN).