BriefGPT.xyz
Sep, 2018
超先验引导的无监督潜在表征分离
Hyperprior Induced Unsupervised Disentanglement of Latent Representations
HTML
PDF
Abdul Fatir Ansari, Harold Soh
TL;DR
本研究提出了一种在深度生成模型中实现无监督的解缠结潜在表示的方法，通过在VAE的潜在空间中引入反Wishart先验来实现潜变量之间的统计独立性，能够在多个数据集上优于beta-VAE和与最先进的FactorVAE相竞争。
Abstract
We address the problem of
unsupervised disentanglement
of latent representations learnt via
deep generative models
. In contrast to current approaches that operate on the evidence lower bound (ELBO), we argue that
→