# Angular-Based Word Meta-Embedding Learning

arXiv: https://arxiv.org/abs/1808.04334 · HTML: https://mlsanity.com/paper/1808.04334/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Aug 2018.

## Built on by (most cited first, 2 of 2)

- [Learning Meta Word Embeddings by Unsupervised Weighted Concatenation of Source Embeddings](https://mlsanity.com/paper/2204.12386/) (2022)
- [Unsupervised Attention-based Sentence-Level Meta-Embeddings from Contextualised Language Models](https://mlsanity.com/paper/2204.07746/) (2022)

## Abstract (arXiv)

Ensembling word embeddings to improve distributed word representations has shown good success for natural language processing tasks in recent years. These approaches either carry out straightforward mathematical operations over a set of vectors or use unsupervised learning to find a lower-dimensional representation. This work compares meta-embeddings trained for different losses, namely loss functions that account for angular distance between the reconstructed embedding and the target and those that account normalized distances based on the vector length. We argue that meta-embeddings are better to treat the ensemble set equally in unsupervised learning as the respective quality of each embedding is unknown for upstream tasks prior to meta-embedding. We show that normalization methods that account for this such as cosine and KL-divergence objectives outperform meta-embedding trained on standard ℓ1 and ℓ2 loss on textitdefacto word similarity and relatedness datasets and find it outperforms existing meta-learning strategies.