Hierarchical Reinforcement Learning With Guidance for Multi-Domain Dialogue Policy.
Mahdin RohmatillahJen-Tzung ChienPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2023)
Keyphrases
- multi domain
- hierarchical reinforcement learning
- spoken dialogue systems
- rbac model
- reward function
- cross domain
- markov decision process
- reinforcement learning
- average reward
- optimal policy
- domain specific
- state abstraction
- model free
- dialogue system
- state space
- role based access control
- markov decision processes
- heterogeneous networks
- reinforcement learning algorithms
- policy iteration
- long run
- general purpose
- semi supervised
- machine learning
- sufficient conditions
- social networks