论文标题
一项研究影响gan可学习性的性状的研究
A study of traits that affect learnability in GANs
论文作者
论文摘要
生成对抗网络gan是使用两个神经网络的算法体系结构，将一个神经网络与相反的相反，以提出可以通过真实数据传递的新的合成数据实例。训练gan是一个具有挑战性的问题，它要求我们应用高级技术，例如高参数调整，建筑工程等。许多不同的损失，正规化和归一化方案，已经提出了网络架构来解决不同类型数据集的挑战性问题。有必要理解实验性观察并推断出简单的理论。在本文中，我们使用参数化的合成数据集执行经验实验，以探测哪些特征会影响学习性。
Generative Adversarial Networks GANs are algorithmic architectures that use two neural networks, pitting one against the opposite so as to come up with new, synthetic instances of data that can pass for real data. Training a GAN is a challenging problem which requires us to apply advanced techniques like hyperparameter tuning, architecture engineering etc. Many different losses, regularization and normalization schemes, network architectures have been proposed to solve this challenging problem for different types of datasets. It becomes necessary to understand the experimental observations and deduce a simple theory for it. In this paper, we perform empirical experiments using parameterized synthetic datasets to probe what traits affect learnability.
