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 2102.10952cs.CLcs.AIcs.LGcs.LO
收藏

关系Tsetlin机及其在自然语言理解中的应用

A Relational Tsetlin Machine with Applications to Natural Language Understanding

Centre for AI Research(人工智能研究中心);University of Agder(阿格德尔大学);University of Pittsburgh(匹兹堡大学)

Rupsa Saha, Ole-Christoffer Granmo, Vladimir I. Zadorozhny, Morten Goodwin

摘要:

TM是一种模式识别方法,它使用有限状态机进行学习,并使用命题逻辑来表示模式。除了原生解释外,它们还为各种任务提供了具有竞争力的准确性。在本文中,我们提出了一个一阶逻辑为基础的框架与Herbrand语义,以提高计算能力的TM。由此产生的TM是相关的,可以利用自然语言中出现的逻辑结构,学习代表现实世界中行为和后果如何相关的规则。结果是Horn子句的逻辑程序,引入了非结构化数据的结构化视图。在闭域问答中,一阶表示产生了10倍更紧凑的KB,同时回答准确率从94.83%提高到99.48%。该方法对错误,缺失和多余的信息更加强大,提取了对现实世界理解重要的文本方面。

英文摘要:

TMs are a pattern recognition approach that uses finite state machines for learning and propositional logic to represent patterns. In addition to being natively interpretable, they have provided competitive accuracy for various tasks. In this paper, we increase the computing power of TMs by proposing a first-order logic-based framework with Herbrand semantics. The resulting TM is relational and can take advantage of logical structures appearing in natural language, to learn rules that represent how actions and consequences are related in the real world. The outcome is a logic program of Horn clauses, bringing in a structured view of unstructured data. In closed-domain question-answering, the first-order representation produces 10x more compact KBs, along with an increase in answering accuracy from 94.83% to 99.48%. The approach is further robust towards erroneous, missing, and superfluous information, distilling the aspects of a text that are important for real-world understanding.

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。

↑