AgentGraph: Towards Universal Dialogue Management with Structured Deep Reinforcement Learning.
Lu ChenZhi ChenBowen TanSishan LongMilica GasicKai YuPublished in: CoRR (2019)
Keyphrases
- dialogue management
- reinforcement learning
- learning agent
- dialogue system
- spoken dialogue systems
- natural language generation
- function approximation
- language understanding
- partially observable markov decision process
- model free
- markov decision processes
- learning algorithm
- optimal policy
- state space
- dynamic programming
- machine learning
- temporal difference
- reinforcement learning algorithms
- natural language processing
- supervised learning
- multi domain
- learning process
- transfer learning
- optimal control
- virtual world
- partially observable