论文标题

VFlow:具有变化数据增强的更具表现力的生成流量

VFlow: More Expressive Generative Flows with Variational Data Augmentation

论文作者

Chen, Jianfei, Lu, Cheng, Chenli, Biqi, Zhu, Jun, Tian, Tian

论文摘要

生成流是用于密度建模的有前途的可拖动模型,该模型定义了具有可逆变换的概率分布。但是,拖延性会对生成流施加架构约束,从而使其比其他类型的生成模型的表现不佳。在这项工作中,我们研究了先前被忽视的约束,即由于可逆性,所有中间表示都必须具有相同的维度与原始数据具有相同的维度,从而限制了网络的宽度。我们通过使用一些额外的维度来增强数据并共同学习以增强数据的生成流以及在变异推理框架下增强维度的分布来解决这一约束。我们的方法Vflow是生成流的概括,因此始终表现更好。与现有的生成流相结合,Vflow在CIFAR-10数据集上每个维度实现了新的2.98位,并且比以前的型号更紧凑,以达到相似的建模质量。

Generative flows are promising tractable models for density modeling that define probabilistic distributions with invertible transformations. However, tractability imposes architectural constraints on generative flows, making them less expressive than other types of generative models. In this work, we study a previously overlooked constraint that all the intermediate representations must have the same dimensionality with the original data due to invertibility, limiting the width of the network. We tackle this constraint by augmenting the data with some extra dimensions and jointly learning a generative flow for augmented data as well as the distribution of augmented dimensions under a variational inference framework. Our approach, VFlow, is a generalization of generative flows and therefore always performs better. Combining with existing generative flows, VFlow achieves a new state-of-the-art 2.98 bits per dimension on the CIFAR-10 dataset and is more compact than previous models to reach similar modeling quality.

扫码加入交流群

加入微信交流群

微信交流群二维码

扫码加入学术交流群,获取更多资源