MLCVLGJul 8, 2020

NVAE: A Deep Hierarchical Variational Autoencoder

arXiv:2007.03898v31096 citationsHas Code
AI Analysis

This work addresses the problem of improving VAE performance for image generation, particularly for researchers and practitioners in generative modeling, though it is incremental as it builds on existing VAE frameworks.

The paper tackled the performance gap of variational autoencoders (VAEs) compared to other generative models by designing a deep hierarchical VAE architecture, achieving state-of-the-art results on image datasets like CIFAR-10 with a reduction from 2.98 to 2.91 bits per dimension and generating high-quality images up to 256x256 pixels.

Normalizing flows, autoregressive models, variational autoencoders (VAEs), and deep energy-based models are among competing likelihood-based frameworks for deep generative learning. Among them, VAEs have the advantage of fast and tractable sampling and easy-to-access encoding networks. However, they are currently outperformed by other models such as normalizing flows and autoregressive models. While the majority of the research in VAEs is focused on the statistical challenges, we explore the orthogonal direction of carefully designing neural architectures for hierarchical VAEs. We propose Nouveau VAE (NVAE), a deep hierarchical VAE built for image generation using depth-wise separable convolutions and batch normalization. NVAE is equipped with a residual parameterization of Normal distributions and its training is stabilized by spectral regularization. We show that NVAE achieves state-of-the-art results among non-autoregressive likelihood-based models on the MNIST, CIFAR-10, CelebA 64, and CelebA HQ datasets and it provides a strong baseline on FFHQ. For example, on CIFAR-10, NVAE pushes the state-of-the-art from 2.98 to 2.91 bits per dimension, and it produces high-quality images on CelebA HQ. To the best of our knowledge, NVAE is the first successful VAE applied to natural images as large as 256$\times$256 pixels. The source code is available at https://github.com/NVlabs/NVAE .

Code Implementations10 repos

Data from Papers with Code (CC-BY-SA-4.0)

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes