Reinforcement Learning with Large Action Spaces for Neural Machine Translation.
Asaf YehudaiLeshem ChoshenLior FoxOmri AbendPublished in: COLING (2022)
Keyphrases
- machine translation
- action space
- reinforcement learning
- state space
- markov decision processes
- state and action spaces
- continuous state
- real valued
- natural language processing
- cross lingual
- information extraction
- function approximators
- action selection
- continuous state spaces
- stochastic processes
- cross language information retrieval
- target language
- natural language
- function approximation
- markov decision process
- finite state
- single agent
- markov decision problems
- machine translation system
- statistical machine translation
- reinforcement learning algorithms
- policy search
- multi agent systems
- ranking algorithm
- word alignment
- knowledge representation
- optimal policy
- text mining
- learning algorithm