Neural Machine TranslationCambridge University Press, 18 июн. 2020 г. - Всего страниц: 406 Deep learning is revolutionizing how machine translation systems are built today. This book introduces the challenge of machine translation and evaluation - including historical, linguistic, and applied context -- then develops the core deep learning methods used for natural language applications. Code examples in Python give readers a hands-on blueprint for understanding and implementing their own machine translation systems. The book also provides extensive coverage of machine learning tricks, issues involved in handling various forms of data, model enhancements, and current challenges and methods for analysis and visualization. Summaries of the current research in the field make this a state-of-the-art textbook for undergraduate and graduate classes, as well as an essential reference for researchers and developers interested in other applications of neural methods in the broader field of human language processing. |
Содержание
The Translation Problem | 3 |
Uses of Machine Translation | 19 |
History | 29 |
Evaluation | 41 |
Neural Networks | 67 |
Computation Graphs | 89 |
Neural Language Models | 103 |
Neural Translation Models | 125 |
Revisiting Words | 213 |
Adaptation | 239 |
Beyond Parallel Corpora | 263 |
Linguistic Structure | 281 |
Current Challenges | 293 |
Analysis and Visualization | 311 |
Bibliography | 343 |
Author Index | 375 |
Другие издания - Просмотреть все
Часто встречающиеся слова и выражения
2018 Conference Association for Computational attention mechanism beam search BLEU score computation graph Computational Linguistics Conference on Empirical Conference on Machine context words decoder Embed Embed Embed embedding spaces Empirical Methods encoder English error feed-forward FF FF FF Figure gradient hidden layer Human Language Technologies hypothesis input sentence input word L2 norm language model language pairs learning rate machine learning machine translation models machine translation systems match matrix Methods in Natural metrics monolingual data multiple n-best list n-gram Natural Language Processing neural language model neural machine translation neural translation model neurons North American Chapter optimization out-of-domain output word parallel corpus parameters problem Proceedings recurrent neural network reference translation relevant RNN RNN RNN self-attention semantic sentence pairs sequence Short Papers Softmax statistical machine translation step subword syntactic target task token training data training examples training objective typically values vector word alignment word embeddings word prediction word representations