RL-NMT: Reinforcement Learning Fine-tuning for Improved Neural Machine Translation of Burmese Dialects.
Ye Kyaw ThuThazin Myint OoThepchai SupnithiPublished in: MMAsia (Workshops) (2023)
Keyphrases
- machine translation
- reinforcement learning
- fine tuning
- fine tuned
- function approximation
- language independent
- cross lingual
- state space
- natural language processing
- model free
- fitted q iteration
- information extraction
- reinforcement learning algorithms
- statistical machine translation
- chinese english
- natural language generation
- target language
- learning algorithm
- language processing
- dynamic programming
- language resources
- natural language
- markov decision processes
- cross language information retrieval
- optimal policy
- word sense disambiguation
- temporal difference
- machine learning
- word alignment
- action space
- machine translation system
- rl algorithms
- function approximators
- transfer learning