相信很多人都遇到过训练一个deep model的过程中,loss突然变成了NaN。在这里对这个问题做一个总结。 一般来说,出现NaN有以下几种情...
2018 · NAACL · BP · Deep Contextualized Word Representations 问题:以往的词向量表示...
2018 · NIPS · Learned in Translation: Contextualized Word Vectors 想法来源:从...
2018 · ICLR · A NEW METHOD OF REGION EMBEDDING FOR TEXT CLASSIFICATION 想...
2018 · ICLR · SKIP RNN: LEARNING TO SKIP STATE UPDATES IN RECURRENT NEUR...
2018· ACL · Unsupervised Learning of Sentence Embeddings using Compositi...
2018 · ACL · Some of Them Can be Guessed! Exploring the Effect of Lingui...
2018 · ACL · LSTMs Exploit Linguistic Attributes of Data 想法来源:实验性质文章。主要是...
2018 · ACL · Exploring Semantic Properties of Sentence Embeddings 想法来源:探...