Comparing user simulations for dialogue strategy learning

Recent studies show that user simulations can be used to generate training corpora for learning dialogue strategies automatically. However, it is unclear what type of simulation is most suitable in a particular task setting. We observe that a simulation which generates random behaviors in a restricted way outperforms simulations that mimic human user behaviors in a statistical way. Our finding suggests that we do not always need to construct a realistic user simulation. Since constructing realistic user simulations is not a trivial task, we can save engineering cost by wisely choosing simulation models that are appropriate for our task. © 2011 ACM