Continual Learning for Task-oriented Dialogue System with Iterative Network Pruning, Expanding and Masking

TL;DR

This paper introduces TPEM, a continual learning method for task-oriented dialogue systems that uses iterative network pruning, expanding, and masking to retain old knowledge while efficiently learning new tasks.

Contribution

The paper presents a novel continual learning approach combining pruning, expanding, and masking to improve dialogue systems' ability to learn sequential tasks without forgetting.

Findings

TPEM outperforms strong baselines on seven tasks across three datasets.

The method effectively preserves old task performance while learning new tasks.

Extensive experiments demonstrate the robustness and efficiency of TPEM.

Abstract

This ability to learn consecutive tasks without forgetting how to perform previously trained problems is essential for developing an online dialogue system. This paper proposes an effective continual learning for the task-oriented dialogue system with iterative network pruning, expanding and masking (TPEM), which preserves performance on previously encountered tasks while accelerating learning progress on subsequent tasks. Specifically, TPEM (i) leverages network pruning to keep the knowledge for old tasks, (ii) adopts network expanding to create free weights for new tasks, and (iii) introduces task-specific network masking to alleviate the negative impact of fixed weights of old tasks on new tasks. We conduct extensive experiments on seven different tasks from three benchmark datasets and show empirically that TPEM leads to significantly improved results over the strong competitors. For reproducibility, we submit the code and data at: https://github.com/siat-nlp/TPEM