VFlow: More Expressive Generative Flows with Variational Data
Augmentation
- URL: http://arxiv.org/abs/2002.09741v2
- Date: Wed, 22 Jul 2020 15:27:12 GMT
- Title: VFlow: More Expressive Generative Flows with Variational Data
Augmentation
- Authors: Jianfei Chen, Cheng Lu, Biqi Chenli, Jun Zhu, Tian Tian
- Abstract summary: tractability imposes architectural constraints on generative flows, making them less expressive than other types of generative models.
We tackle this constraint by augmenting the data with some extra dimensions and jointly learning a generative flow for augmented data.
Our approach, VFlow, is a generalization of generative flows and therefore always performs better.
- Score: 33.431861316434706
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: 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.
Related papers
- Distribution-Aware Data Expansion with Diffusion Models [55.979857976023695]
We propose DistDiff, a training-free data expansion framework based on the distribution-aware diffusion model.
DistDiff consistently enhances accuracy across a diverse range of datasets compared to models trained solely on original data.
arXiv Detail & Related papers (2024-03-11T14:07:53Z) - Guided Flows for Generative Modeling and Decision Making [55.42634941614435]
We show that Guided Flows significantly improves the sample quality in conditional image generation and zero-shot text synthesis-to-speech.
Notably, we are first to apply flow models for plan generation in the offline reinforcement learning setting ax speedup in compared to diffusion models.
arXiv Detail & Related papers (2023-11-22T15:07:59Z) - Generative Flows with Invertible Attentions [135.23766216657745]
We introduce two types of invertible attention mechanisms for generative flow models.
We exploit split-based attention mechanisms to learn the attention weights and input representations on every two splits of flow feature maps.
Our method provides invertible attention modules with tractable Jacobian determinants, enabling seamless integration of it at any positions of the flow-based models.
arXiv Detail & Related papers (2021-06-07T20:43:04Z) - Variational Mixture of Normalizing Flows [0.0]
Deep generative models, such as generative adversarial networks autociteGAN, variational autoencoders autocitevaepaper, and their variants, have seen wide adoption for the task of modelling complex data distributions.
Normalizing flows have overcome this limitation by leveraging the change-of-suchs formula for probability density functions.
The present work overcomes this by using normalizing flows as components in a mixture model and devising an end-to-end training procedure for such a model.
arXiv Detail & Related papers (2020-09-01T17:20:08Z) - SurVAE Flows: Surjections to Bridge the Gap between VAEs and Flows [78.77808270452974]
SurVAE Flows is a modular framework for composable transformations that encompasses VAEs and normalizing flows.
We show that several recently proposed methods, including dequantization and augmented normalizing flows, can be expressed as SurVAE Flows.
arXiv Detail & Related papers (2020-07-06T13:13:22Z) - Normalizing Flows Across Dimensions [10.21537170623373]
We introduce noisy injective flows (NIF), a generalization of normalizing flows that can go across dimensions.
NIF explicitly map the latent space to a learnable manifold in a high-dimensional data space using injective transformations.
Empirically, we demonstrate that a simple application of our method to existing flow architectures can significantly improve sample quality and yield separable data embeddings.
arXiv Detail & Related papers (2020-06-23T14:47:18Z) - SoftFlow: Probabilistic Framework for Normalizing Flow on Manifolds [15.476426879806134]
Flow-based generative models are composed of invertible transformations between two random variables of the same dimension.
In this paper, we propose SoftFlow, a probabilistic framework for training normalizing flows on manifold.
We experimentally show that SoftFlow can capture the innate structure of the manifold data and generate high-quality samples.
We apply the proposed framework to 3D point clouds to alleviate the difficulty of forming thin structures for flow-based models.
arXiv Detail & Related papers (2020-06-08T13:56:07Z) - Normalizing Flows with Multi-Scale Autoregressive Priors [131.895570212956]
We introduce channel-wise dependencies in their latent space through multi-scale autoregressive priors (mAR)
Our mAR prior for models with split coupling flow layers (mAR-SCF) can better capture dependencies in complex multimodal data.
We show that mAR-SCF allows for improved image generation quality, with gains in FID and Inception scores compared to state-of-the-art flow-based models.
arXiv Detail & Related papers (2020-04-08T09:07:11Z) - Closing the Dequantization Gap: PixelCNN as a Single-Layer Flow [16.41460104376002]
We introduce subset flows, a class of flows that can transform finite volumes and allow exact computation of likelihoods for discrete data.
We identify ordinal discrete autoregressive models, including WaveNets, PixelCNNs and Transformers, as single-layer flows.
We demonstrate state-of-the-art results on CIFAR-10 for flow models trained with dequantization.
arXiv Detail & Related papers (2020-02-06T22:58:51Z) - Semi-Supervised Learning with Normalizing Flows [54.376602201489995]
FlowGMM is an end-to-end approach to generative semi supervised learning with normalizing flows.
We show promising results on a wide range of applications, including AG-News and Yahoo Answers text data.
arXiv Detail & Related papers (2019-12-30T17:36:33Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.