Optimizing Policy via Deep Reinforcement Learning for Dialogue Management.
Guanghao XuHyungjung LeeMyoung-Wan KooJungyun SeoPublished in: BigComp (2018)
Keyphrases
- dialogue management
- reinforcement learning
- partially observable markov decision process
- optimal policy
- partially observable
- learning agent
- action selection
- dialogue system
- state space
- markov decision process
- spoken dialogue systems
- markov decision problems
- partially observable markov decision processes
- language understanding
- decision theoretic
- markov decision processes
- action space
- reward function
- function approximators
- natural language generation
- function approximation
- policy iteration
- reinforcement learning algorithms
- decision problems
- multi agent
- learning algorithm
- virtual humans
- belief state
- model free
- dynamic programming
- machine learning
- solving problems
- long run
- supervised learning
- finite state
- markov chain
- optimal control
- learning tasks