Trajectory-Based Meta-Learning for Out-Of-Vocabulary Word Embedding Learning

TL;DR

This paper introduces Leap, a trajectory-based meta-learning algorithm for out-of-vocabulary word embedding, which improves stability and performance over MAML by leveraging the entire learning trajectory.

Contribution

The paper proposes Leap, a novel meta-learning algorithm that uses the full learning trajectory to enhance OOV word embedding, addressing MAML's instability issues.

Findings

Leap performs comparably or better than MAML on benchmark datasets.

The choice of contexts significantly influences OOV embedding quality.

Trajectory-based meta-learning improves stability and results in OOV embedding learning.

Abstract

Word embedding learning methods require a large number of occurrences of a word to accurately learn its embedding. However, out-of-vocabulary (OOV) words which do not appear in the training corpus emerge frequently in the smaller downstream data. Recent work formulated OOV embedding learning as a few-shot regression problem and demonstrated that meta-learning can improve results obtained. However, the algorithm used, model-agnostic meta-learning (MAML) is known to be unstable and perform worse when a large number of gradient steps are used for parameter updates. In this work, we propose the use of Leap, a meta-learning algorithm which leverages the entire trajectory of the learning process instead of just the beginning and the end points, and thus ameliorates these two issues. In our experiments on a benchmark OOV embedding learning dataset and in an extrinsic evaluation, Leap performs comparably or better than MAML. We go on to examine which contexts are most beneficial to learn an OOV embedding from, and propose that the choice of contexts may matter more than the meta-learning employed.