CLASSIC: Continual and Contrastive Learning of Aspect Sentiment Classification Tasks
CLASSIC: Continual and Contrastive Learning of Aspect Sentiment Classification Tasks
复制标题
DOI：
10.18653/v1/2021.emnlp-main.550
复制
发表时间：
2021-12
期刊：
影响因子：
--
通讯作者：
Zixuan Ke;Bing Liu;Hu Xu;Lei Shu
中科院分区：
文献类型：
--
作者：
Zixuan Ke;Bing Liu;Hu Xu;Lei Shu
本文研究领域增量式学习(DIL)环境下一系列方面情感分类任务的连续学习。每个任务来自不同的域或产品。DIL设置特别适合ASC，因为在测试中，系统不需要知道测试数据所属的任务/域。据我们所知，以前没有人研究过ASC的这种设置。本文提出了一个新的模型，称为经典模型。关键的新颖性是一种对比性的持续学习方法，它既实现了任务之间的知识转移，又实现了从旧任务到新任务的知识蒸馏，从而在测试中消除了对任务ID的需要。实验结果表明，CLASSIC算法具有较高的有效性。
This paper studies continual learning (CL) of a sequence of aspect sentiment classification (ASC) tasks in a particular CL setting called domain incremental learning (DIL). Each task is from a different domain or product. The DIL setting is particularly suited to ASC because in testing the system needs not know the task/domain to which the test data belongs. To our knowledge, this setting has not been studied before for ASC. This paper proposes a novel model called CLASSIC. The key novelty is a contrastive continual learning method that enables both knowledge transfer across tasks and knowledge distillation from old tasks to the new task, which eliminates the need for task ids in testing. Experimental results show the high effectiveness of CLASSIC.