Multimodal Word Distributions

最新推荐文章于 2022-02-15 13:32:28 发布

算法学习者

最新推荐文章于 2022-02-15 13:32:28 发布

阅读量996

点赞数

分类专栏： paper reading NLP

paper reading 同时被 2 个专栏收录

85 篇文章 0 订阅

订阅专栏

NLP

43 篇文章 5 订阅

订阅专栏

Multimodal Word Distributions

Ben Athiwaratkun, Andrew Gordon Wilson

(Submitted on 27 Apr 2017)

Word embeddings provide point representations of words containing useful semantic information. We introduce multimodal word distributions formed from Gaussian mixtures, for multiple word meanings, entailment, and rich uncertainty information. To learn these distributions, we propose an energy-based max-margin objective. We show that the resulting approach captures uniquely expressive semantic information, and outperforms alternatives, such as word2vec skip-grams, and Gaussian embeddings, on benchmark datasets such as word similarity and entailment.

Comments:	This paper also appears at ACL 2017
Subjects:	Machine Learning (stat.ML); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Learning (cs.LG)
Cite as:	arXiv:1704.08424 [stat.ML]
	(or arXiv:1704.08424v1 [stat.ML] for this version)