MaGNET: Uniform Sampling from Deep Generative Network Manifolds Without Retraining

10/15/2021
by   Ahmed Imtiaz Humayun, et al.
9

Deep Generative Networks (DGNs) are extensively employed in Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and their variants to approximate the data manifold, and data distribution on that manifold. However, training samples are often obtained based on preferences, costs, or convenience producing artifacts in the empirical data distribution e.g., the large fraction of smiling faces in the CelebA dataset or the large fraction of dark-haired individuals in FFHQ. These inconsistencies will be reproduced when sampling from the trained DGN, which has far-reaching potential implications for fairness, data augmentation, anomaly detection, domain adaptation, and beyond. In response, we develop a differential geometry based sampler – coined MaGNET – that, given any trained DGN, produces samples that are uniformly distributed on the learned manifold. We prove theoretically and empirically that our technique produces a uniform distribution on the manifold regardless of the training set distribution. We perform a range of experiments on various datasets and DGNs. One of them considers the state-of-the-art StyleGAN2 trained on FFHQ dataset, where uniform sampling via MaGNET increases distribution precision and recall by 4.1 41.2

READ FULL TEXT

page 1

page 9

page 18

page 19

page 20

page 21

page 22

page 24

research
02/24/2021

Self-Diagnosing GAN: Diagnosing Underrepresented Samples in Generative Adversarial Networks

Despite remarkable performance in producing realistic samples, Generativ...
research
10/16/2019

Optimal Transport Based Generative Autoencoders

The field of deep generative modeling is dominated by generative adversa...
research
08/19/2023

Deep Generative Modeling-based Data Augmentation with Demonstration using the BFBT Benchmark Void Fraction Datasets

Deep learning (DL) has achieved remarkable successes in many disciplines...
research
01/21/2023

AQuaMaM: An Autoregressive, Quaternion Manifold Model for Rapidly Estimating Complex SO(3) Distributions

Accurately modeling complex, multimodal distributions is necessary for o...
research
10/09/2020

Unsupervised 3D Brain Anomaly Detection

Anomaly detection (AD) is the identification of data samples that do not...
research
12/16/2019

MimicGAN: Robust Projection onto Image Manifolds with Corruption Mimicking

In the past few years, generative models like Generative Adversarial Net...
research
06/04/2019

Encoding Invariances in Deep Generative Models

Reliable training of generative adversarial networks (GANs) typically re...

Please sign up or login with your details

Forgot password? Click here to reset