StarSpace: Embed All The Things!

Ledell Wu, Adam Fisch, Sumit Chopra, Keith Adams, Antoine Bordes +1 more
9/12/2017
cs.CL

Official code: facebookresearch/StarSpace (3,955 stars, MIT, last commit 2022-12-04).

Abstract

We present StarSpace, a general-purpose neural embedding model that can solve a wide variety of problems: labeling tasks such as text classification, ranking tasks such as information retrieval/web search, collaborative filtering-based or content-based recommendation, embedding of multi-relational graphs, and learning word, sentence or document level embeddings. In each case the model works by embedding those entities comprised of discrete features and comparing them against each other -- learning similarities dependent on the task. Empirical results on a number of tasks show that StarSpace is highly competitive with existing methods, whilst also being generally applicable to new cases where those methods are not.

View on arXivView PDF

Code Implementations(1)

facebookresearch/StarSpaceOfficial100%
MIT

Learning embeddings for classification, retrieval and ranking.

3,955523C++, CMakeJun 28, 20173 years agoMIT

Similar Papers

Survey on Embedding Models for Knowledge Graph and its Applications

85% match2024

Siamese CBOW: Optimizing Word Embeddings for Sentence Representations

83% match2016

Representation Learning for Attributed Multiplex Heterogeneous Network

82% match1 repo5382019

How to Generate a Good Word Embedding?

82% match1 repo2382015

TabEmbed: Benchmarking and Learning Generalist Embeddings for Tabular Understanding

82% match1 repo52026

Improving Interpretability via Explicit Word Interaction Graph Layer

82% match1 repo32023

Cite this paper

@article{wu2017starspace,
  title  = {StarSpace: Embed All The Things!},
  author = {Ledell Wu and Adam Fisch and Sumit Chopra and Keith Adams and Antoine Bordes and Jason Weston},
  year   = {2017},
  eprint = {1709.03856},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1709.03856}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1709.03856.svg)](https://codeofpaper.com/paper/1709.03856)

Discussion