Ninth International Conference on Spoken Language Processing

Pittsburgh, PA, USA
September 17-21, 2006

Rapid Simulation-Driven Reinforcement Learning of Multimodal Dialog Strategies in Human-Robot Interaction

Thomas Prommer, Hartwig Holzapfel, Alex Waibel

Universitšt Karlsruhe, Germany

In this work we propose a procedure model for rapid automatic strategy learning in multimodal dialogs. Our approach is tailored for typical task-oriented human-robot dialog interactions, with no prior knowledge about the expected user and system dynamics being present. For such scenarios, we propose the use of stochastic dialog simulation for strategy learning, where the user and system error models are solely trained through the initial execution of an inexpensive Wizard-of-Oz experiment. We argue that for the addressed dialogs, already a small data corpus combined with a low-conditioned simulation model facilitates learning of strong and complex dialog strategies. To validate our overall approach, we empirically show the supremacy of the learned strategy over a hand-crafted strategy for a concrete human-robot dialog scenario. To the authorsí knowledge, this work is the first to perform strategy learning from multimodal dialog simulation.

Full Paper

Bibliographic reference.  Prommer, Thomas / Holzapfel, Hartwig / Waibel, Alex (2006): "Rapid simulation-driven reinforcement learning of multimodal dialog strategies in human-robot interaction", In INTERSPEECH-2006, paper 1551-Wed2FoP.4.