Consistent Alignment of World Embedding Models

Abstract

Word embedding models offer continuous vector representations that can capture rich contextual semantics based on their word co-occurrence patterns. While these word vectors can provide very effective features used in many NLP tasks such as clustering similar words and inferring learning relationships, many challenges and open research questions remain. In this paper, we propose a solution that aligns variations of the same model (or different models) in a joint low-dimensional latent space leveraging carefully generated synthetic data points. This generative process is inspired by the observation that a variety of linguistic relationships is captured by simple linear operations in embedded space. We demonstrate that our approach can lead to substantial improvements in recovering quality embeddings of local neighborhoods aligned and fused across different input word models.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 02, 2017
Accession Number
AD1028485

Entities

People

  • Brandon Oselio
  • Cem S. Sahin
  • Rajmonda S. Caceres
  • William M. Campbell

Organizations

  • MIT Lincoln Laboratory

Tags

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Artificial Intelligence Software
  • Computational Science
  • Dimensionality Reduction
  • Embedding
  • Engineering
  • Information Processing
  • Information Systems
  • Language
  • Linguistics
  • Models
  • Natural Language Processing
  • Natural Languages
  • Neural Networks
  • Training
  • Vocabulary
  • Workshops

Fields of Study

  • Computer science

Readers

  • Computational Linguistics
  • Computer Vision.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Machine Translation
  • Space