%0 Generic %1 yin2015learning %A Yin, Wenpeng %A Schütze, Hinrich %D 2015 %K embedding learning proposal relatedness semantic tau %T Learning Meta-Embeddings by Using Ensembles of Embedding Sets %U http://arxiv.org/abs/1508.04257 %X Word embeddings -- distributed representations of words -- in deep learning are beneficial for many tasks in natural language processing (NLP). However, different embedding sets vary greatly in quality and characteristics of the captured semantics. Instead of relying on a more advanced algorithm for embedding learning, this paper proposes an ensemble approach of combining different public embedding sets with the aim of learning meta-embeddings. Experiments on word similarity and analogy tasks and on part-of-speech tagging show better performance of meta-embeddings compared to individual embedding sets. One advantage of meta-embeddings is the increased vocabulary coverage. We will release our meta-embeddings publicly.

@misc{yin2015learning, abstract = {Word embeddings -- distributed representations of words -- in deep learning are beneficial for many tasks in natural language processing (NLP). However, different embedding sets vary greatly in quality and characteristics of the captured semantics. Instead of relying on a more advanced algorithm for embedding learning, this paper proposes an ensemble approach of combining different public embedding sets with the aim of learning meta-embeddings. Experiments on word similarity and analogy tasks and on part-of-speech tagging show better performance of meta-embeddings compared to individual embedding sets. One advantage of meta-embeddings is the increased vocabulary coverage. We will release our meta-embeddings publicly.}, added-at = {2016-11-28T09:50:12.000+0100}, author = {Yin, Wenpeng and Schütze, Hinrich}, biburl = {https://www.bibsonomy.org/bibtex/2763a9dc183e7e605b2cfbb8970001f14/thoni}, interhash = {cc52e493511fd5a66a147077db32687c}, intrahash = {763a9dc183e7e605b2cfbb8970001f14}, keywords = {embedding learning proposal relatedness semantic tau}, note = {cite arxiv:1508.04257Comment: 10 pages, 6 figures}, timestamp = {2016-11-28T15:30:27.000+0100}, title = {Learning Meta-Embeddings by Using Ensembles of Embedding Sets}, url = {http://arxiv.org/abs/1508.04257}, year = 2015 }