How to Build User Simulators to Train RL-based Dialog Systems
This work addresses a key bottleneck in developing RL-based dialog systems for researchers and practitioners, offering a standardized evaluation approach, though it is incremental in nature.
The paper tackles the challenge of building effective user simulators for training reinforcement learning-based dialog systems by proposing a standardized method and evaluating six simulators with automatic metrics and human assessments, providing a comprehensive framework for comparing their quality and impact on trained systems.
User simulators are essential for training reinforcement learning (RL) based dialog models. The performance of the simulator directly impacts the RL policy. However, building a good user simulator that models real user behaviors is challenging. We propose a method of standardizing user simulator building that can be used by the community to compare dialog system quality using the same set of user simulators fairly. We present implementations of six user simulators trained with different dialog planning and generation methods. We then calculate a set of automatic metrics to evaluate the quality of these simulators both directly and indirectly. We also ask human users to assess the simulators directly and indirectly by rating the simulated dialogs and interacting with the trained systems. This paper presents a comprehensive evaluation framework for user simulator study and provides a better understanding of the pros and cons of different user simulators, as well as their impacts on the trained systems.