A User Simulator for Task-Completion Dialogues

TL;DR

This paper introduces a new, publicly available user simulator for task-oriented dialogues in the movie-booking domain, facilitating reinforcement learning research without extensive data collection.

Contribution

The paper presents a hybrid rule-based and data-driven user simulator for movie-booking dialogues, supporting multiple tasks and easy integration of new agents.

Findings

The simulator effectively models movie-booking dialogues.

Agents trained on the simulator perform well in tasks.

The framework simplifies testing and comparison of dialogue agents.

Abstract

Despite widespread interests in reinforcement-learning for task-oriented dialogue systems, several obstacles can frustrate research and development progress. First, reinforcement learners typically require interaction with the environment, so conventional dialogue corpora cannot be used directly. Second, each task presents specific challenges, requiring separate corpus of task-specific annotated data. Third, collecting and annotating human-machine or human-human conversations for task-oriented dialogues requires extensive domain knowledge. Because building an appropriate dataset can be both financially costly and time-consuming, one popular approach is to build a user simulator based upon a corpus of example dialogues. Then, one can train reinforcement learning agents in an online fashion as they interact with the simulator. Dialogue agents trained on these simulators can serve as an effective starting point. Once agents master the simulator, they may be deployed in a real environment to interact with humans, and continue to be trained online. To ease empirical algorithmic comparisons in dialogues, this paper introduces a new, publicly available simulation framework, where our simulator, designed for the movie-booking domain, leverages both rules and collected data. The simulator supports two tasks: movie ticket booking and movie seeking. Finally, we demonstrate several agents and detail the procedure to add and test your own agent in the proposed framework.