Curriculum Learning Based on Reward Sparseness for Deep Reinforcement Learning of Task Completion Dialogue Management.
Atsushi SaitoPublished in: SCAI@EMNLP (2018)
Keyphrases
- reinforcement learning
- dialogue management
- learning process
- learning agent
- learning algorithm
- state space
- dynamic programming
- function approximation
- action selection
- reinforcement learning algorithms
- supervised learning
- natural language processing
- domain independent
- temporal difference
- learning mechanism
- learning capabilities
- e learning
- machine learning