Gromov-Wasserstein Alignment of Word Embedding Spaces

By David Alvarez-Melis, Tommi S. Jaakkola. Conference on Empirical Methods in Natural Language Processing, 2018

359 citations recorded by the source corpus.

Cross-lingual or cross-domain correspondences play key roles in tasks ranging from machine translation to transfer learning. Recently, purely unsupervised methods operating on monolingual embeddings have become effective alignment tools. Current state-of-the-art methods, however, involve multiple steps, including heuristic post-hoc refinement strategies. In this paper, we cast the correspondence problem directly as an optimal transport (OT) problem, building on the idea that word embeddings arise from metric recovery algorithms. Indeed, we exploit the Gromov-Wasserstein distance that measures how similarities between pairs of words relate across languages. We show that our OT objective can be estimated efficiently, requires little or no tuning, and results in performance comparable with the state-of-the-art in various unsupervised word translation tasks.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/16b221c831855766dff661e88d0a7a9e835a6cb0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Graph Optimal Transport for Cross-Domain Alignment (2020, methodology): “The solver for GWD can be readily developed based on Algorithm 1, where p , q are deﬁned as uniform distributions (as shown in Algorithm 2), following Alvarez-Melis & Jaakkola (2018).”
A Survey of Cross-lingual Word Embedding Models (2017, background): “Alvarez-Melis and Jaakkola (2018) solve an optimal transport in order to learn an alignment between the monolingual word embedding spaces.”
Blackbox Meets Blackbox: Representational Similarity & Stability Analysis of Neural Language Models and Brains (2019, background): “RSA can also be applied to deep learning models (Laakso and Cottrell, 2000; Dharmaretnam and Fyshe, 2018; Alvarez-Melis and Jaakkola, 2018; Wang et al., 2018; Chrupała and Alishahi, 2019).”
Gromov-Wasserstein Learning for Graph Matching and Node Embedding (2019, background): “Besides graphics and computer vision, recently its potential for other applications has been investigated, e.g., matching vocabulary sets between different languages (Alvarez-Melis & Jaakkola, 2018) and matching weighted directed networks (Chowdhury & Mémoli, 2018).”
Unsupervised Parallel Sentence Extraction with Parallel Segment Detection Helps Machine Translation (2019, methodology): “Others used word neighborhood information to create an initial mapping (Artetxe et al., 2018a; Alvarez-Melis and Jaakkola, 2018).”
Bilingual Lexicon Induction through Unsupervised Machine Translation (2019, methodology): “At the same time, several recent methods have formulated embedding mappings as an optimal transport problem (Zhang et al., 2017b; Grave et al., 2018; Alvarez-Melis and Jaakkola, 2018).”