VFlow: More Expressive Generative Flows with Variational Data Augmentation

ICML 2020  ·  Jianfei Chen, Cheng Lu, Biqi Chenli, Jun Zhu, Tian Tian ·

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. read more

PDF Abstract ICML 2020 PDF

Datasets


Results from the Paper


Ranked #20 on Image Generation on CIFAR-10 (bits/dimension metric)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Generation CIFAR-10 VFlow bits/dimension 2.98 # 20

Methods


No methods listed for this paper. Add relevant methods here