arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.03569cs.CLcs.LGcs.NEstat.ML
收藏

基于神经高斯Copula的变分自编码器

Neural Gaussian Copula for Variational Autoencoder

University of California Santa Barbara(加州大学圣巴巴拉分校)

Prince Zizhuang Wang, William Yang Wang

摘要:

变分语言模型寻求用近似的变分后验估计潜变量的后验。该模型通常假设变分后验被分解,即使真正的后验不是。在此假设下学习的变分后验并没有捕捉到潜在变量之间的依赖关系。我们认为,这将导致一个典型的训练问题,称为后崩溃观察到所有其他变语言模型。我们提出高斯Copula变分自编码器(VAE)来避免这个问题。Copula被广泛用于高维随机变量的相关性和依赖性建模,因此它有助于维护VAE中丢失的依赖关系。实证结果表明,通过使用神经参数copula显式地对潜变量的相关性进行建模,我们可以避免这种训练困难,同时获得与所有其他VAE方法相比具有竞争力的结果。

英文摘要:

Variational language models seek to estimate the posterior of latent variables with an approximated variational posterior. The model often assumes the variational posterior to be factorized even when the true posterior is not. The learned variational posterior under this assumption does not capture the dependency relationships over latent variables. We argue that this would cause a typical training problem called posterior collapse observed in all other variational language models. We propose Gaussian Copula Variational Autoencoder (VAE) to avert this problem. Copula is widely used to model correlation and dependencies of high-dimensional random variables, and therefore it is helpful to maintain the dependency relationships that are lost in VAE. The empirical results show that by modeling the correlation of latent variables explicitly using a neural parametric copula, we can avert this training difficulty while getting competitive results among all other VAE approaches.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑