Improving Sample-Efficiency in Reinforcement Learning for Dialogue Systems by Using Trainable-Action-Mask.
Yen-Chen WuBo-Hsiang TsengCarl Edward RasmussenPublished in: ICASSP (2020)
Keyphrases
- dialogue system
- reinforcement learning
- dialogue management
- human computer
- spoken dialogue systems
- action selection
- natural language generation
- natural language
- tutorial dialogue
- mixed initiative
- function approximation
- general purpose
- multi agent
- dialogue games
- machine learning
- action space
- transfer learning
- language understanding
- hidden markov models