Gaussian Process based Deep Dyna-Q approach for Dialogue Policy Learning

34026

Knowledge Triples

Source Entity	Relationship	Target Entity
Deep Dyna Q Method	improves	Dialog Policy Learning Task
Deep Dyna Q Method	integrates	Gaussian Process Method
Deep Dyna Q Method	uses	World Model Method
Dialog Policy Learning Task	measures	Task Completion Success Rate Metric
World Model Method	evaluates	Kullback-Leibler Divergence Metric