Findings from Experiments of On-line Joint Reinforcement Learning of Semantic Parser and Dialogue Manager with real Users

TL;DR

This paper explores online joint reinforcement learning of semantic parser and dialogue manager in dialogue systems, demonstrating effective training with few dialogues and analyzing challenges in maintaining consistent training strategies.

Contribution

It introduces variants of simultaneous online learning for dialogue systems and provides experimental insights into their effectiveness and training challenges.

Findings

Good performance achieved with only a few hundred dialogues

Variants of online learning can surpass handcrafted systems

Training consistency remains a key challenge

Abstract

Design of dialogue systems has witnessed many advances lately, yet acquiring huge set of data remains an hindrance to their fast development for a new task or language. Besides, training interactive systems with batch data is not satisfactory. On-line learning is pursued in this paper as a convenient way to alleviate these difficulties. After the system modules are initiated, a single process handles data collection, annotation and use in training algorithms. A new challenge is to control the cost of the on-line learning borne by the user. Our work focuses on learning the semantic parsing and dialogue management modules (speech recognition and synthesis offer ready-for-use solutions). In this context we investigate several variants of simultaneous learning which are tested in user trials. In our experiments, with varying merits, they can all achieve good performance with only a few hundreds of training dialogues and overstep a handcrafted system. The analysis of these experiments gives us some insights, discussed in the paper, into the difficulty for the system's trainers to establish a coherent and constant behavioural strategy to enable a fast and good-quality training phase.