A Discriminative Latent-Variable Model for Bilingual Lexicon Induction

08/28/2018
by   Sebastian Ruder, et al.
2

We introduce a novel discriminative latent-variable model for the task of bilingual lexicon induction. Our model combines the bipartite matching dictionary prior of Haghighi et al. (2008) with a state-of-the-art embedding-based approach. To train the model, we derive an efficient Viterbi EM algorithm. We provide empirical improvements on six language pairs under two metrics and show that the prior theoretically and empirically helps to mitigate the hubness problem. We also demonstrate how previous work may be viewed as a similarly fashioned latent-variable model, albeit with a different prior.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/11/2018

Conditional Variational Autoencoder for Neural Machine Translation

We explore the performance of latent variable models for conditional tex...
research
12/17/2022

Latent Variable Representation for Reinforcement Learning

Deep latent variable models have achieved significant empirical successe...
research
06/17/2019

Open Domain Event Extraction Using Neural Latent Variable Models

We consider open domain event extraction, the task of extracting unconst...
research
06/17/2013

Spectral Experts for Estimating Mixtures of Linear Regressions

Discriminative latent-variable models are typically learned using EM or ...
research
11/01/2018

Latent Visual Cues for Neural Machine Translation

In this work, we propose to model the interaction between visual and tex...
research
09/21/2017

SpectralFPL: Online Spectral Learning for Single Topic Models

This paper studies how to efficiently learn an optimal latent variable m...
research
06/27/2023

Exploiting Inferential Structure in Neural Processes

Neural Processes (NPs) are appealing due to their ability to perform fas...

Please sign up or login with your details

Forgot password? Click here to reset