Modelling Sentence Pairs via Reinforcement Learning: An Actor-Critic Approach to Learn the Irrelevant Words.
Mahtab AhmedRobert E. MercerPublished in: AAAI (2020)
Keyphrases
- actor critic
- reinforcement learning
- temporal difference
- policy gradient
- sentence pairs
- reinforcement learning algorithms
- optimal control
- approximate dynamic programming
- function approximation
- word pairs
- gradient method
- neuro fuzzy
- function approximators
- model free
- learning agent
- parallel corpus
- policy iteration
- machine learning
- keywords
- rl algorithms
- optimal policy
- word level
- state space
- n gram
- average reward
- markov decision processes
- machine translation system
- word recognition
- word segmentation
- semi supervised
- action selection
- step size
- learning tasks
- transfer learning