Autonomous Sub-domain Modeling for Dialogue Policy with Hierarchical Deep Reinforcement Learning.
Giovanni Yoko KristiantoHuiwen ZhangBin TongMakoto IwayamaYoshiyuki KobayashiPublished in: SCAI@EMNLP (2018)
Keyphrases
- reinforcement learning
- optimal policy
- action selection
- partially observable domains
- policy search
- domain specific
- autonomous learning
- action space
- approximate dynamic programming
- domain independent
- markov decision processes
- control policy
- markov decision process
- dynamic programming
- partially observable
- learning algorithm
- partially observable environments
- human machine
- complex domains
- reinforcement learning algorithms
- neural network
- model free
- function approximation
- transfer learning
- function approximators
- average reward
- cross domain
- domain experts
- multi agent
- state space