Generative Latent Flow

TL;DR

Generative Latent Flow (GLF) is a novel generative modeling approach combining auto-encoders and normalizing flows to improve density matching, convergence speed, and sample quality, outperforming many existing models.

Contribution

The paper introduces GLF, a new model that explicitly maps latent distributions to noise, avoiding over regularization and achieving state-of-the-art results among AE-based models.

Findings

Achieves state-of-the-art sample quality among AE-based models

Demonstrates fast convergence and single-stage training

Competitive with GAN benchmarks

Abstract

In this work, we propose the Generative Latent Flow (GLF), an algorithm for generative modeling of the data distribution. GLF uses an Auto-encoder (AE) to learn latent representations of the data, and a normalizing flow to map the distribution of the latent variables to that of simple i.i.d noise. In contrast to some other Auto-encoder based generative models, which use various regularizers that encourage the encoded latent distribution to match the prior distribution, our model explicitly constructs a mapping between these two distributions, leading to better density matching while avoiding over regularizing the latent variables. We compare our model with several related techniques, and show that it has many relative advantages including fast convergence, single stage training and minimal reconstruction trade-off. We also study the relationship between our model and its stochastic counterpart, and show that our model can be viewed as a vanishing noise limit of VAEs with flow prior. Quantitatively, under standardized evaluations, our method achieves state-of-the-art sample quality among AE based models on commonly used datasets, and is competitive with GANs' benchmarks.