条件概率图产生式对抗网络

产生式对抗网络(generative adversarial networks,简称GANs)可以生成逼真的图像,因此最近被广泛研究.值得注意的是,概率图生成对抗网络(graphical-GAN)将贝叶斯网络引入产生式对抗网络框架,以无监督的方式学习到数据的隐藏结构.提出了条件概率图生成对抗网络(conditional graphical-GAN),它可以在弱监督环境下,利用粗粒度监督信息来学习到更精细而复杂的结构.条件概率图生成对抗网络的推理和学习遵循与graphical-GAN类似的方法.提出了条件概率图生成对抗网络的两个实例.条件高斯混合模型(conditional Gaussian mixture GAN,简称cGMGAN)可以在给出粗粒度标签的情况下从混合数据中学习细粒度聚类.条件状态空间模型(conditional state space GAN,简称cSSGAN)可以在给定对象标签的情况下学习具有多个对象的视频的动态过程.

Conditional Graphical Generative Adversarial Networks

Geneartive adverarial networks (GAN) have been promise on generating realistic images and hence have been studied widely. Notabaly, graphical generative adversarial networks (Graphical-GAN) introduce Bayesian networks to the GAN framework to learn the underlying structures of data in an unsupervised manner. In this paper, we propose a conditional version of Graphical-GAN, which can leverage coarse side information to enhance the Graphical-GAN and learn more fine and complex structures, in weakly-supervised learning settings. The inference and learning of conditional Graphical-GAN follows a similar protocol to Graphical-GAN. We present two instances of conditional Graphical-GAN. The conditional Gaussian mixture GAN can learn fine clusters from mixture data given a corse label. The conditional state space GAN can learn the dynamics of videos with multiple objects given the labels of the objects.