Cascaded LSTMs Based Deep Reinforcement Learning for Goal-Driven Dialogue.
Yue MaXiaojie WangZhenjiang DongHong ChenPublished in: NLPCC (2017)
Keyphrases
- goal driven
- reinforcement learning
- software product line
- service composition
- function approximation
- human machine
- model free
- state space
- face detection
- learning algorithm
- semantic web services
- optimal policy
- markov decision processes
- dialogue system
- mixed initiative
- machine learning
- multi agent
- lightweight
- domain knowledge
- learning process
- expert systems
- database systems