A Bandit Approach to Posterior Dialog Orchestration Under a Budget.
Sohini UpadhyayMayank AgarwalDjallel BouneffoufYasaman KhazaeniPublished in: CoRR (2019)
Keyphrases
- natural language
- posterior distribution
- probability distribution
- web services
- spoken dialog
- mixed initiative
- user interface
- random sampling
- posterior probability
- bandit problems
- spoken dialog systems
- multi armed bandit
- dialog systems
- probabilistic model
- machine learning
- gaussian process
- conversational agent
- database
- information systems
- conversational agents
- limited memory
- data streams
- technology enhanced learning
- regret bounds
- budget constraints
- case study
- decision making
- social network analysis
- learning processes
- bayesian framework