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