Abstract not found

Designing the dialogue policy of a spoken dialogue system involves many nontrivial choices. This paper presents a reinforcement learning approach for automatically optimizing a dialogue policy, which addresses the technical challenges in applying reinforcement learning to a working dialogue system with human users. We report on the design, construction and empirical evaluation of NJFun, an experimental spoken dialogue system that provides users with access to information about fun things to do in New Jersey. Our results show that by optimizing its performance via reinforcement learning, NJFun measurably improves system performance.

This paper describes a novel method by which a spoken dialogue system can learn to choose an optimal dialogue strategy from its experience interacting with human users. The method is

This paper describes a novel method by which a dia-logue agent can learn to choose an optimal dialogue strategy. While it is widely agreed that dialogue strategies should be formulated in terms of com-municative intentions, there has been little work on automatically optimizing an agent's choices when there are multiple ways to realize a communica-tive intention. Our method is based on a combina-tion of learning algorithms and empirical evaluation techniques. The learning component of our method is based on algorithms for reinforcement learning, such as dynamic programming and Q-learning. The empirical component uses the PARADISE evalua-tion framework (Walker et al., 1997) to identify the important peribrmance factors and to provide the performance function needed by the learning algo-rithm. We illustrate our method with a dialogue agent named ELVIS (EmaiL Voice Interactive Sys-tem), that supports access to email over the phone. We show how ELVIS can learn to choose among alternate strategies for agent initiative, for reading messages, and for summarizing email folders. 1

Automatic speech dialogue systems are becoming common. In order to assess their performance, a large sample of real dialogues has to be collected and evaluated. This process is expensive, labor intensive, and prone to errors. To alleviate this situation we propose a user simulation to conduct dialogues with the system under investigation. Using stochastic modeling of real users we can both debug and evaluate a speech dialogue system while it is still in the lab, thus substantially reducing the amount of field testing with real users. 1 Introduction Recent literature shows an increasing number of speech dialogue systems being implemented and used in the field [1, 2, 3, 4, 5, 6]. However, the development of such dialogue systems (especially the dialogue manager) is still considered art rather than an engineering task. While the optimality criteria of a low level component like the speech recognizer are obvious (i.e. generate an accurate transliteration of a speech signal) there exists n...

Collaborative learning tasks involve interaction between multiple participants, who thus need to maintain some degree of mutual understanding. The process by which this is accomplished is termed grounding. The way in which collaboration, grounding and learning take place is largely determined by the task, the situation and the tools available. This paper discusses relations between grounding, collaboration and learning, drawing on research from two main areas: the Language Sciences and Cultural-Historical Activity Theory ("CHAT"). We build a unifying perspective of mutual understanding mediated by material and semiotic tools that can be used for analysis as well as for design of collaborative learning tasks, especially those that are carried out via computer-mediated communication. We illustrate the perspective with reference to a particular computermediated collaborative learning situation in the domain of physics. 1. Introduction Collaborative learning is a complex phenomenon that c...

Recent technological advances have made it possible to build real-time, interactive spoken dialogue systems for a wide variety of applications. However, when users do not respect the limitations of such systems, performance typically degrades. Although users differ with respect to their knowledge of system limitations, and although different dialogue strategies make system limitations more apparent to users, most current systems do not try to improve performance by adapting dialogue behavior to individual users. This paper presents an empirical evaluation of TOOT, an adaptable spoken dialogue system for retrieving train schedules on the web. We conduct an experiment in which 20 users carry out 4 tasks with both adaptable and non-adaptable versions of TOOT, resulting in a corpus of 80 dialogues. The values for a wide range of evaluation measures are then extracted from this corpus. Our results show that adaptable TOOT generally outperforms non-adaptable TOOT, and that the utility of adaptation depends on TOOT's initial dialogue strategies.

This paper describes AUTOSEM, a robust semantic interpretation framework that can operate both at parse time and repair time. The evaluation demon- strates that AUTOSEM achieves a high level of robustness efficiently and without requiring any hand coded knowledge dedicated to repair.

Abstract. Spoken dialogue system performance canvary widely for different users, as well for the same user during different dialogues.This paper presents the design and evaluation of an adaptive version of TOOT, a spoken dialogue system for retrieving online train schedules. Based on rules learned from a set of training dialogues, adaptive TOOT constructs a user model representing whether the user is having speech recognition problems as a particular dialogue progresses. Adaptive TOOT then automatically adapts its dialogue strategies based on this dynamically changing user model. An empirical evaluation of the system demonstrates the utility of the approach. Key words. adaptive spoken dialogue systems, hypothesis testing for the e¡ectiveness of adaptations, PARADISE for evaluating performance measures, speech recognition, user model acquisition via machine learning 1.

Designing the dialogue strategy of a spokeu dialogue system involves mauy noutrivial choices. This paper presents a reinforcement learning approach for automatically optimizing a dialogue strategy that addresses the technical challenges in applying reinforcement learning to a working dialogue system with hmnan users. We then show that our approach measurably improves performance in an experimental system.