MaGNET: Uniform Sampling from Deep Generative Network Manifolds Without Retraining
MaGNET: Uniform Sampling from Deep Generative Network Manifolds Without Retraining
复制标题
DOI：
--
复制
发表时间：
2021-10
期刊：
影响因子：
--
通讯作者：
Ahmed Imtiaz Humayun;Randall Balestriero;Richard Baraniuk
中科院分区：
文献类型：
--
作者：
Ahmed Imtiaz Humayun;Randall Balestriero;Richard Baraniuk
深层生成网络（DGN）广泛用于生成对抗网络（GAN），变异自动编码器（VAE）及其变体中，以近似数据歧管和分布。但是，培训样本通常由于成本或收集方便而在多种方式上以不均匀的方式分配。例如，Celeba数据集包含很大一部分的微笑面孔。从训练有素的DGN采样时，这些不一致将被复制，例如公平或数据增强。作为回应，我们开发了磁铁，这是一种新颖且理论上动机的潜在空间采样器，用于任何预训练的DGN，它会产生在学习的歧管上均匀分布的样品。我们在各种数据集和DGN上执行一系列实验，例如，对于在FFHQ数据集中训练的最先进的stylegan2，通过磁铁进行均匀采样可提高分布精度，并召回4.1 \％\％\＆3.0 \％，并减少和减少。性别偏见为41.2 \％，无需标签或再培训。由于均匀的分布并不意味着统一的语义分布，因此我们还分别探索了在磁铁采样下产生样品的语义属性如何变化。
Deep Generative Networks (DGNs) are extensively employed in Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and their variants to approximate the data manifold and distribution. However, training samples are often distributed in a non-uniform fashion on the manifold, due to costs or convenience of collection. For example, the CelebA dataset contains a large fraction of smiling faces. These inconsistencies will be reproduced when sampling from the trained DGN, which is not always preferred, e.g., for fairness or data augmentation. In response, we develop MaGNET, a novel and theoretically motivated latent space sampler for any pre-trained DGN, that produces samples uniformly distributed on the learned manifold. We perform a range of experiments on various datasets and DGNs, e.g., for the state-of-the-art StyleGAN2 trained on FFHQ dataset, uniform sampling via MaGNET increases distribution precision and recall by 4.1\% \&3.0\% and decreases gender bias by 41.2\%, without requiring labels or retraining. As uniform distribution does not imply uniform semantic distribution, we also explore separately how semantic attributes of generated samples vary under MaGNET sampling.