Rewarding Smatch: Transition-Based AMR Parsing with Reinforcement Learning.
Tahira NaseemAbhishek ShahHui WanRadu FlorianSalim RoukosMiguel BallesterosPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- transition model
- function approximation
- natural language
- natural language processing
- state space
- reinforcement learning algorithms
- markov decision processes
- learning algorithm
- unsupervised learning
- optimal policy
- temporal difference
- robotic control
- autonomous learning
- partially observable
- model free
- pattern matching
- dynamic programming
- machine learning
- optimal control
- learning classifier systems
- state transition
- transfer learning
- language model
- finite element analysis
- information retrieval
- natural language parsing
- neural network