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 1709.10367cs.CLcs.LGstat.ML
收藏

分组数据的结构化嵌入模型

Structured Embedding Models for Grouped Data

Maja Rudolph, Francisco Ruiz, Susan Athey, David Blei

摘要:

词嵌入是分析语言的强大方法,指数族嵌入(EFE)将其扩展到其他类型的数据。在这里,我们开发了结构化指数族嵌入(S-EFE),这是一种用于发现在相关数据组中变化的嵌入的方法。我们研究了美国国会演讲的词汇使用如何在各州和党派之间变化,ArXiv各部分的词汇使用如何不同,以及杂货的共同购买模式如何在季节之间变化。我们的方法成功的关键是这些组共享统计信息。我们开发了两种共享策略:分层建模和摊销。我们证明了这种方法的好处,在实证研究的演讲,摘要和购物篮。我们展示了S-EFE如何使特定于组的解释词的使用,并优于EFE在预测举行了数据。

英文摘要:

Word embeddings are a powerful approach for analyzing language, and exponential family embeddings (EFE) extend them to other types of data. Here we develop structured exponential family embeddings (S-EFE), a method for discovering embeddings that vary across related groups of data. We study how the word usage of U.S. Congressional speeches varies across states and party affiliation, how words are used differently across sections of the ArXiv, and how the co-purchase patterns of groceries can vary across seasons. Key to the success of our method is that the groups share statistical information. We develop two sharing strategies: hierarchical modeling and amortization. We demonstrate the benefits of this approach in empirical studies of speeches, abstracts, and shopping baskets. We show how S-EFE enables group-specific interpretation of word usage, and outperforms EFE in predicting held-out data.

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。

↑