Continual Learning in Task-Oriented Dialogue Systems

Andrea Madotto,Zhiguang Wang,Bing Liu,Seungwhan Moon,Paul Crook,Eunjoon Cho,Zhou Yu,Zhenpeng Zhou,Pascale Fung,Zhaojiang Lin

doi:10.18653/v1/2021.emnlp-main.590

Abstract

Continual learning in task-oriented dialogue systems can allow us to add new domains and functionalities through time without incurring the high cost of a whole system retraining. In this paper, we propose a continual learning benchmark for task-oriented dialogue systems with 37 domains to be learned continuously in four settings, such as intent recognition, state tracking, natural language generation, and end-to-end. Moreover, we implement and compare multiple existing continual learning baselines, and we propose a simple yet effective architectural method based on residual adapters. Our experiments demonstrate that the proposed architectural method and a simple replay-based strategy perform comparably well but they both achieve inferior performance to the multi-task learning baseline, in where all the data are shown at once, showing that continual learning in task-oriented dialogue systems is a challenging task. Furthermore, we reveal several trade-offs between different continual learning methods in term of parameter usage and memory size, which are important in the design of a task-oriented dialogue system. The proposed benchmark is released together with several baselines to promote more research in this direction.

Highlights

Task-oriented dialogue systems (ToDs) are the core technology of the current state-of-the-art smart assistants (e.g. Alexa, Siri, Portal, etc.)
We report several trade-offs in terms of parameter usage, memory size and training time, which are important in the design of a task-oriented dialogue system
The following syntax is used: 1. We propose a benchmark for continual learning in ToDs, with 37 tasks to be learned con

Summary

Introduction

Task-oriented dialogue systems (ToDs) are the core technology of the current state-of-the-art smart assistants (e.g. Alexa, Siri, Portal, etc.). These systems are either modularized as a pipeline of multiple components, namely, natural language understanding (NLU), dialogue state tracking (DST), dialogue policy (DP) and natural language generation (NLG), or end-to-end, where a single model implicitly learns how to issue APIs and system responses. In the CL setting the main challenge is catastrophic forgetting (McCloskey and Cohen, 1989) This phenomena happens because there is a distributional shift between the tasks in the curriculum which leads to catastrophic forgetting of the previously acquired knowledge. To overcome this challenge three kinds of methods are usually deployed: loss regularization, for avoiding interference with the previously learned tasks, rehearsal, which use episodic memory to recall previously learned tasks, and architectural, which add task-specific parame-

Objectives

Methods

Results