English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Word embedding learning methods require a large number of occurrences of aword to accurately learn its embedding. However, out-of-vocabulary (OOV) wordswhich do not appear in the training corpus emerge frequently in the smallerdownstream data. Recent work formulated OOV embedding learning as a few-shotregression problem and demonstrated that meta-learning can improve resultsobtained. However, the algorithm used, model-agnostic meta-learning (MAML) isknown to be unstable and perform worse when a large number of gradient stepsare used for parameter updates. In this work, we propose the use of Leap, ameta-learning algorithm which leverages the entire trajectory of the learningprocess instead of just the beginning and the end points, and thus amelioratesthese two issues. In our experiments on a benchmark OOV embedding learningdataset and in an extrinsic evaluation, Leap performs comparably or better thanMAML. We go on to examine which contexts are most beneficial to learn an OOVembedding from, and propose that the choice of contexts may matter more thanthe meta-learning employed.

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