Character n-Gram Embeddings to Improve RNN Language Models
2019 ◽
Vol 33
◽
pp. 5074-5082
◽
Keyword(s):
This paper proposes a novel Recurrent Neural Network (RNN) language model that takes advantage of character information. We focus on character n-grams based on research in the field of word embedding construction (Wieting et al. 2016). Our proposed method constructs word embeddings from character ngram embeddings and combines them with ordinary word embeddings. We demonstrate that the proposed method achieves the best perplexities on the language modeling datasets: Penn Treebank, WikiText-2, and WikiText-103. Moreover, we conduct experiments on application tasks: machine translation and headline generation. The experimental results indicate that our proposed method also positively affects these tasks
Keyword(s):
2017 ◽
Vol 7
(7)
◽
pp. 444
Keyword(s):
Keyword(s):
2017 ◽
Vol 108
(1)
◽
pp. 271-282
◽
2018 ◽
pp. 368-377
◽
Keyword(s):
2013 ◽
Vol 21
(2)
◽
pp. 201-226
◽
2014 ◽
Vol 102
(1)
◽
pp. 81-92
◽
Keyword(s):
2020 ◽
Vol 34
(05)
◽
pp. 8766-8774
◽