arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1812.07546cs.CL
收藏

面向个性化领域分类的有监督领域使能注意

Supervised Domain Enablement Attention for Personalized Domain Classification

Amazon Alexa(亚马逊Alexa)

Joo-Kyung Kim, Young-Bum Kim

摘要:

在用于自然语言理解的大规模域分类中,利用每个用户的域启用信息(其指用户的优选或认证域)与注意力机制已经被示出为提高整体域分类性能。在本文中,我们提出了一个监督使能注意力机制,它利用sigmoid激活的注意力加权,使注意力可以计算具有更大的表达能力,而没有权重和约束的softmax注意力。通过监督注意力,明确鼓励注意力权重与地面实况的独热向量的对应元素相似,并且通过自蒸馏利用其他启用域的注意力信息。通过对大规模IPDA的实际话语进行评估,我们表明我们的方法显着提高了域分类性能。

英文摘要:

In large-scale domain classification for natural language understanding, leveraging each user's domain enablement information, which refers to the preferred or authenticated domains by the user, with attention mechanism has been shown to improve the overall domain classification performance. In this paper, we propose a supervised enablement attention mechanism, which utilizes sigmoid activation for the attention weighting so that the attention can be computed with more expressive power without the weight sum constraint of softmax attention. The attention weights are explicitly encouraged to be similar to the corresponding elements of the ground-truth's one-hot vector by supervised attention, and the attention information of the other enabled domains is leveraged through self-distillation. By evaluating on the actual utterances from a large-scale IPDA, we show that our approach significantly improves domain classification performance.

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。

↑