Reinforcement Learning for Bandit Neural Machine Translation with Simulated Human Feedback.
Khanh NguyenHal Daumé IIIJordan L. Boyd-GraberPublished in: EMNLP (2017)
Keyphrases
- machine translation
- reinforcement learning
- language independent
- information extraction
- sensory inputs
- cross lingual
- natural language processing
- natural language generation
- target language
- word sense disambiguation
- human subjects
- language processing
- statistical machine translation
- natural language
- word alignment
- machine translation system
- language resources
- cross language information retrieval
- state space
- word level
- parallel corpora
- multilingual documents
- machine learning
- query translation
- dynamic programming
- knowledge base
- data mining
- tasks in natural language processing
- statistical translation models