Domain Adaptive Dialog Generation via Meta Learning

TL;DR

This paper introduces DAML, a meta-learning based dialog system that efficiently adapts to new domains with minimal data, reducing the need for extensive annotation and training.

Contribution

The paper presents a novel end-to-end dialog system using meta-learning to enable rapid domain adaptation with few training samples.

Findings

Achieved state-of-the-art performance on simulated dialog datasets.

Demonstrated effective generalization to new dialog domains.

Enabled efficient adaptation with minimal training data.

Abstract

Domain adaptation is an essential task in dialog system building because there are so many new dialog tasks created for different needs every day. Collecting and annotating training data for these new tasks is costly since it involves real user interactions. We propose a domain adaptive dialog generation method based on meta-learning (DAML). DAML is an end-to-end trainable dialog system model that learns from multiple rich-resource tasks and then adapts to new domains with minimal training samples. We train a dialog system model using multiple rich-resource single-domain dialog data by applying the model-agnostic meta-learning algorithm to dialog domain. The model is capable of learning a competitive dialog system on a new domain with only a few training examples in an efficient manner. The two-step gradient updates in DAML enable the model to learn general features across multiple tasks. We evaluate our method on a simulated dialog dataset and achieve state-of-the-art performance, which is generalizable to new tasks.