Reusing Discriminators for Encoding: Towards Unsupervised Image-to-Image Translation

NICE-GAN: 把判别器重用为编码器的图像翻译模型.

最优传输(Optimal Transport)问题与Wasserstein距离

Wasserstein Distance.

Softmax GAN

把生成对抗网络建模为Softmax函数.

On Convergence and Stability of GANs

DRAGAN：调整梯度惩罚的插值空间.

Invertible Residual Networks

i-ResNet：通过变分解量化和结构设计改进流模型.

Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design

Flow++：通过变分解量化和结构设计改进流模型.