arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.08571cs.LGcs.CL
收藏

基于离散自回归变分注意力模型的文本建模

Discrete Auto-regressive Variational Attention Models for Text Modeling

Harbin Institute of Technology, Shenzhen(哈尔滨工业大学(深圳));The Chinese University of Hong Kong(香港中文大学)

Xianghong Fang, Haoli Bai, Jian Li, Zenglin Xu, Michael Lyu, Irwin King

摘要:

变分自编码器(VAE)已广泛应用于文本建模。然而,在实践中,他们受到两个挑战的困扰:信息代表性不足和后验崩溃。前者的出现是因为只有LSTM编码器的最后一个隐藏状态被转换到潜在空间中,这通常不足以总结数据。后者是VAE训练过程中的一个长期存在的问题,因为优化被困在灾难性的局部最优。在本文中,我们提出了离散自回归变分注意力模型(DAVAM),以解决这一挑战。具体来说,我们引入了一个自回归变分注意方法,以丰富的潜在空间,有效地捕捉语义依赖的输入。我们进一步设计了离散的潜在空间的变分注意力和数学证明,我们的模型是免费的后验崩溃。语言建模任务的大量实验表明DAVAM对几个VAE同行的优越性。

英文摘要:

Variational autoencoders (VAEs) have been widely applied for text modeling. In practice, however, they are troubled by two challenges: information underrepresentation and posterior collapse. The former arises as only the last hidden state of LSTM encoder is transformed into the latent space, which is generally insufficient to summarize the data. The latter is a long-standing problem during the training of VAEs as the optimization is trapped to a disastrous local optimum. In this paper, we propose Discrete Auto-regressive Variational Attention Model (DAVAM) to address the challenges. Specifically, we introduce an auto-regressive variational attention approach to enrich the latent space by effectively capturing the semantic dependency from the input. We further design discrete latent space for the variational attention and mathematically show that our model is free from posterior collapse. Extensive experiments on language modeling tasks demonstrate the superiority of DAVAM against several VAE counterparts.

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。

↑