Creswell Antonia, Bharath Anil Anthony
IEEE Trans Neural Netw Learn Syst. 2018 Nov 2. doi: 10.1109/TNNLS.2018.2875194.
Generative adversarial networks (GANs) learn a deep generative model that is able to synthesize novel, high-dimensional data samples. New data samples are synthesized by passing latent samples, drawn from a chosen prior distribution, through the generative model. Once trained, the latent space exhibits interesting properties that may be useful for downstream tasks such as classification or retrieval. Unfortunately, GANs do not offer an ``inverse model,'' a mapping from data space back to latent space, making it difficult to infer a latent representation for a given data sample. In this paper, we introduce a technique, inversion, to project data samples, specifically images, to the latent space using a pretrained GAN. Using our proposed inversion technique, we are able to identify which attributes of a data set a trained GAN is able to model and quantify GAN performance, based on a reconstruction loss. We demonstrate how our proposed inversion technique may be used to quantitatively compare the performance of various GAN models trained on three image data sets. We provide codes for all of our experiments in the website (https://github.com/ToniCreswell/InvertingGAN).
生成对抗网络（GAN）学习一种深度生成模型，该模型能够合成新颖的高维数据样本。通过将从选定的先验分布中抽取的潜在样本输入生成模型，从而合成新的数据样本。一旦训练完成，潜在空间会展现出一些有趣的特性，这些特性可能对诸如分类或检索等下游任务有用。不幸的是，GAN没有提供一个“逆模型”，即从数据空间到潜在空间的映射，这使得为给定的数据样本推断潜在表示变得困难。在本文中，我们介绍了一种名为反演的技术，使用预训练的GAN将数据样本（特别是图像）投影到潜在空间。使用我们提出的反演技术，我们能够基于重建损失确定训练好的GAN能够建模数据集的哪些属性，并量化GAN的性能。我们展示了如何使用我们提出的反演技术来定量比较在三个图像数据集上训练的各种GAN模型的性能。我们在网站（https://github.com/ToniCreswell/InvertingGAN）上提供了所有实验的代码。
