Box-To-Box Transformations for Modeling Joint Hierarchies
Box-To-Box Transformations for Modeling Joint Hierarchies
复制标题
用于建模联合层次结构的框到框转换
DOI：
10.18653/v1/2021.repl4nlp-1.28
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
A. McCallum
中科院分区：
文献类型：
--
作者：
S. Dasgupta;Xiang Li;Michael Boratko;Dongxu Zhang;A. McCallum
结构化知识库中实体和关系的学习表示是一个活跃的研究领域，非常强调选择适当的几何来捕获例如ISA或HasPart关系中利用的层次结构。框嵌入(Vilny等人，2018；Li等人，2019；Dasgupta等人，2020)将概念表示为n维超矩形，当在传递闭包的子集上进行训练时，能够嵌入层次结构。在Patel等人(2020)中，作者证明了只需要传递约简，并通过用新节点扩充图来进一步扩展盒嵌入来捕获联合层次。虽然用这种方法可以表示联合层次，但每个层次的参数是解耦的，使得层次之间的泛化是不可行的。在这项工作中，我们引入了一种学习的盒到盒转换，它尊重每个层次的结构。我们证明，这不仅提高了跨层次组合边建模的能力，而且能够从传递约简的子集进行泛化。
Learning representations of entities and relations in structured knowledge bases is an active area of research, with much emphasis placed on choosing the appropriate geometry to capture the hierarchical structures exploited in, for example, isa or haspart relations. Box embeddings (Vilnis et al., 2018; Li et al., 2019; Dasgupta et al., 2020), which represent concepts as n-dimensional hyperrectangles, are capable of embedding hierarchies when training on a subset of the transitive closure. In Patel et al., (2020), the authors demonstrate that only the transitive reduction is required and further extend box embeddings to capture joint hierarchies by augmenting the graph with new nodes. While it is possible to represent joint hierarchies with this method, the parameters for each hierarchy are decoupled, making generalization between hierarchies infeasible. In this work, we introduce a learned box-to-box transformation that respects the structure of each hierarchy. We demonstrate that this not only improves the capability of modeling cross-hierarchy compositional edges but is also capable of generalizing from a subset of the transitive reduction.
影响因子：
4.1
作者：
NOSOFSKY, RM
通讯作者：
NOSOFSKY, RM