NVAE: A Deep Hierarchical Variational Autoencoder

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... (read more)

PDF Abstract NeurIPS 2020 PDF NeurIPS 2020 Abstract

Results from the Paper


Ranked #2 on Image Generation on FFHQ 256 x 256 (bits/dimension metric)

     Get a GitHub badge
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Image Generation CelebA 256x256 NVAE w/ flow bpd 0.70 # 3
Image Generation CIFAR-10 NVAE w/ flow bits/dimension 2.91 # 8
Image Generation FFHQ 256 x 256 NVAE w/ flow bits/dimension 0.69 # 2
Image Generation ImageNet 32x32 NVAE w/ flow bpd 3.92 # 6

Methods used in the Paper