This paper makes progress on several open theoretical issues related to Generative Adversarial Networks. A definition is provided for what it means for the training to generalize, and it is shown that generalization is not guaranteed for the popular distances between distributions such as Jensen-Shannon or Wasserstein. We introduce a new metric called neural net distance for which generalization does occur. We also show that an approximate pure equilibrium in the 2-player game exists for a natural training objective (Wasserstein). Showing such a result has been an open problem (for any training objective). Finally, the above theoretical ideas lead us to propose a new training protocol, MIX+GAN, which can be combined with any existing method. We present experiments showing that it stabilizes and improves some existing methods.

通过对生成对抗网络的训练，我们发现它可能没有良好的泛化特性，并且展示了训练成功不代表训练的分布接近目标分布，但是泛化确实存在于弱度量的神经网络距离中。我们证明了当生成器容量和训练集大小适中时，鉴别器/生成器博弈存在近似的纯平衡，这启发了MIX+GAN协议的存在可以与任何现有的GAN训练相结合，并在实践中改善了其中一些。

生成对抗网络中的泛化和均衡