Mathematical Formula Representation via Tree Embeddings

Mathematical Formula Representation via Tree Embeddings
复制标题

DOI:
--
复制
发表时间:
2021
期刊:
影响因子:
4.6
通讯作者:
Zichao Wang;Andrew S. Lan;Richard Baraniuk
Zichao Wang;Andrew S. Lan;Richard Baraniuk
中科院分区:
综合性期刊3区
文献类型:
--
作者:
Zichao Wang;Andrew S. Lan;Richard Baraniuk

文献摘要

被引文献

相似文献

我们提出了一个新的框架,学习数学公式表示使用树嵌入。通过将每个符号公式(如数学方程)表示为运算符树,我们可以显式地捕获其固有的结构和语义属性。我们的框架包括一个树编码器,它将公式的运算符树编码成一个向量,和一个树解码器,它从运算符树格式的向量生成一个公式。为了提高公式树生成的质量,我们开发了一种具有独立科学价值的新型树束搜索算法。我们验证了我们的框架上的公式重建任务和类似的公式检索任务上的一个新的真实世界的数据集超过770k公式在线收集。我们的实验结果表明,我们的框架显着优于各种基线。
We propose a new framework for learning mathematical formula representations using tree embeddings. By representing each symbolic formula (such as math equation) as an operator tree , we can explicitly capture its inherent structural and semantic properties. Our framework consists of a tree encoder that encodes the formula’s operator tree into a vector and a tree decoder that generates a formula from a vector in operator tree format. To improve the quality of formula tree generation, we develop a novel tree beam search algorithm that is of independent scientific interest. We validate our framework on a formula reconstruction task and a similar formula retrieval task on a new real-world dataset of over 770k formulae collected online. Our experimental results show that our framework significantly outperforms various baselines.