BriefGPT.xyz
Mar, 2019
诊断和优化VAE模型
Diagnosing and Enhancing VAE Models
HTML
PDF
Bin Dai, David Wipf
TL;DR
本研究针对变分自编码器（VAEs）的效率和各种假设，提出了新的VAE增强方案，该方案不需要任何超参数或敏感调整，并且与多种GAN模型竞争时，能产生清晰的样本和稳定的FID分数。
Abstract
Although variational autoencoders (
vaes
) represent a widely influential deep generative model, many aspects of the underlying
energy function
remain poorly understood. In particular, it is commonly believed that
→