Continual Learning Based on Sub-Networks and Task SimilarityDownload PDF

Published: 01 Feb 2023, Last Modified: 13 Feb 2023Submitted to ICLR 2023Readers: Everyone
Keywords: Continual learning, NLP tasks, Task Similarity, Sub-network
TL;DR: A continual learning method based on sub-networks and task similarity is proposed and evaluated on NLP classification, generation, and extraction problems.
Abstract: Continual learning (CL) has two main objectives: preventing catastrophic forgetting (CF) and encouraging knowledge transfer (KT) across tasks. The existing literature mainly tries to overcome CF. Although some papers have focused on both CF and KT, they may still suffer from CF because of their ineffective handling of previous tasks and/or poor task similarity detection mechanisms to achieve KT. This work presents a new CL method that addresses the above issues. First, it overcomes CF by isolating the knowledge of each task via a learned mask that indicates a sub-network. Second, it proposes a novel technique to compute how important each mask is to the new task, which indicates how the new task is similar to an underlying old task. Similar tasks can share the same mask/subnetwork for KT, while dissimilar tasks use different masks/sub-networks for CF prevention. Comprehensive experiments have been conducted using a range of NLP problems, including classification, generation, and extraction to show that the proposed method consistently outperforms prior state-of-the-art baselines.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics
Submission Guidelines: Yes
Please Choose The Closest Area That Your Submission Falls Into: Applications (eg, speech processing, computer vision, NLP)
Supplementary Material: zip
15 Replies

Loading