We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: Learning Mathematical Properties of Integers

Abstract: Embedding words in high-dimensional vector spaces has proven valuable in many natural language applications. In this work, we investigate whether similarly-trained embeddings of integers can capture concepts that are useful for mathematical applications. We probe the integer embeddings for mathematical knowledge, apply them to a set of numerical reasoning tasks, and show that by learning the representations from mathematical sequence data, we can substantially improve over number embeddings learned from English text corpora.
Comments: BlackboxNLP 2021
Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
Cite as: arXiv:2109.07230 [cs.CL]
  (or arXiv:2109.07230v1 [cs.CL] for this version)

Submission history

From: Maria Ryskina [view email]
[v1] Wed, 15 Sep 2021 11:54:28 GMT (5798kb,D)

Link back to: arXiv, form interface, contact.