Offline Reinforcement Learning for Mixture-of-Expert Dialogue Management.
Dhawal GuptaYinlam ChowAzamat TulepbergenovMohammad GhavamzadehCraig BoutilierPublished in: NeurIPS (2023)
Keyphrases
- dialogue management
- reinforcement learning
- learning agent
- dialogue system
- spoken dialogue systems
- natural language generation
- function approximation
- language understanding
- partially observable markov decision process
- reinforcement learning algorithms
- learning algorithm
- state space
- multi agent
- partially observable
- optimal policy
- mixed initiative
- virtual humans
- learning capabilities
- machine learning
- model free
- human experts
- dynamic programming
- natural language
- knowledge engineers
- domain experts