In this Chapter we summarise and discuss the main arguments for the use of RL in dialogue development, and introduce the reader to the technical details of this approach. We provide the reader with more background knowledge about the basic principles and ideas behind RL, and we relate these principles to basic characteristics of dialogue interaction.
KeywordsReinforcement Learn Belief State Dialogue System Dialogue Strategy Partially Observable Markov Decision Process
Unable to display preview. Download preview PDF.