Đang chuẩn bị nút TẢI XUỐNG, xin hãy chờ
Tải xuống
Over several years, we have developed an approach to spoken dialogue systems that includes rule-based and trainable dialogue managers, spoken language understanding and generation modules, and a comprehensive dialogue system architecture. We present a Reinforcement Learning-based dialogue system that goes beyond standard rule-based models and computes on-line decisions of the best dialogue moves. The key concept of this work is that we bridge the gap between manually written dialog models (e.g. rule-based) and adaptive computational models such as Partially Observable Markov Decision Processes (POMDP) based dialogue managers. . | Combining POMDPs trained with User Simulations and Rule-based Dialogue Management in a Spoken Dialogue System Sebastian Varges Silvia Quarteroni Giuseppe Riccardi Alexei V. Ivanov Pierluigi Roberti Department of Information Engineering and Computer Science University of Trento 38050 Povo di Trento Italy varges silviaq riccardi ivanov roberti @disi.unitn.it Abstract Over several years we have developed an approach to spoken dialogue systems that includes rule-based and trainable dialogue managers spoken language understanding and generation modules and a comprehensive dialogue system architecture. We present a Reinforcement Learning-based dialogue system that goes beyond standard rule-based models and computes on-line decisions of the best dialogue moves. The key concept of this work is that we bridge the gap between manually written dialog models e.g. rule-based and adaptive computational models such as Partially Observable Markov Decision Processes POMDP based dialogue managers. 1 Reinforcement Learning-based Dialogue Management In recent years Machine Learning techniques in particular Reinforcement Learning RL have been applied to the task of dialogue management DM Levin et al. 2000 Williams and Young 2006 . A major motivation is to improve robustness in the face of uncertainty for example due to speech recognition errors. A further motivation is to improve adaptivity w.r.t. different user behaviour and application recognition environments. The Reinforcement Learning framework is attractive because it offers a statistical model representing the dynamics of the interaction between system and user. This is in contrast to the supervised learning approach of learning system behaviour based on a fixed corpus Higashinaka et al. 2003 . To explore the range of dialogue management strategies a simulation environment is required that includes a simulated user Schatz-mann et al. 2006 if one wants to avoid the prohibitive cost of using human subjects. We demonstrate the .
