Actor-Double-Critic: Incorporating Model-Based Critic for Task-Oriented Dialogue Systems

Abstract

In order to improve the sample-efficiency of deep reinforcement learning (DRL), we implemented imagination augmented agent (I2A) in spoken dialogue systems (SDS). Although I2A achieves a higher success rate than baselines by augmenting predicted future into a policy network, its complicated architecture introduces unwanted instability. In this work, we propose actor-double-critic (ADC) to improve the stability and overall performance of I2A. ADC simplifies the architecture of I2A to reduce excessive parameters and hyperparameters. More importantly, a separate model-based critic shares parameters between actions and makes back-propagation explicit.In our experiments on Cambridge Restaurant Booking task, ADC enhances success rates considerably and shows robustness to imperfect environment models. In addition, ADC exhibits the stability and sample-efficiency as significantly reducing the baseline standard deviation of success rates and reaching the 80% success rate with half training data.

Task-based dialogue policy learning based on decision transformer

“…Model-free reinforcement learning methods interact directly with pre-built environments or real users to learn dialogue policies [4]. Model-based reinforcement learning is comprised of two simultaneous learning modules: model learning and policy learning [5].…”