Adapting BERT for Continual Learning of a Sequence of Aspect Sentiment Classification Tasks

TL;DR

This paper introduces B-CL, a capsule network-based model that enables continual learning for aspect sentiment classification, effectively transferring knowledge and preventing forgetting across sequential tasks.

Contribution

It presents a novel capsule network approach for continual learning in aspect sentiment classification, addressing knowledge transfer and retention issues.

Findings

B-CL significantly improves performance on new and old ASC tasks.

The model effectively transfers knowledge forward and backward.

Extensive experiments validate B-CL's effectiveness.

Abstract

This paper studies continual learning (CL) of a sequence of aspect sentiment classification (ASC) tasks. Although some CL techniques have been proposed for document sentiment classification, we are not aware of any CL work on ASC. A CL system that incrementally learns a sequence of ASC tasks should address the following two issues: (1) transfer knowledge learned from previous tasks to the new task to help it learn a better model, and (2) maintain the performance of the models for previous tasks so that they are not forgotten. This paper proposes a novel capsule network based model called B-CL to address these issues. B-CL markedly improves the ASC performance on both the new task and the old tasks via forward and backward knowledge transfer. The effectiveness of B-CL is demonstrated through extensive experiments.