Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2404.14721 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909271584669696 |
|---|---|
| author | Hong, Chenxing Jin, Yan Kang, Zhiqi Chen, Yizhou Li, Mengke Lu, Yang Wang, Hanzi |
| author_facet | Hong, Chenxing Jin, Yan Kang, Zhiqi Chen, Yizhou Li, Mengke Lu, Yang Wang, Hanzi |
| contents | Existing continual learning literature relies heavily on a strong assumption that tasks arrive with a balanced data stream, which is often unrealistic in real-world applications. In this work, we explore task-imbalanced continual learning (TICL) scenarios where the distribution of task data is non-uniform across the whole learning process. We find that imbalanced tasks significantly challenge the capability of models to control the trade-off between stability and plasticity from the perspective of recent prompt-based continual learning methods. On top of the above finding, we propose Dynamically Anchored Prompting (DAP), a prompt-based method that only maintains a single general prompt to adapt to the shifts within a task stream dynamically. This general prompt is regularized in the prompt space with two specifically designed prompt anchors, called boosting anchor and stabilizing anchor, to balance stability and plasticity in TICL. Remarkably, DAP achieves this balance by only storing a prompt across the data stream, therefore offering a substantial advantage in rehearsal-free CL. Extensive experiments demonstrate that the proposed DAP results in 4.5% to 15% absolute improvements over state-of-the-art methods on benchmarks under task-imbalanced settings. Our code is available at https://github.com/chenxing6666/DAP |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_14721 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Dynamically Anchored Prompting for Task-Imbalanced Continual Learning Hong, Chenxing Jin, Yan Kang, Zhiqi Chen, Yizhou Li, Mengke Lu, Yang Wang, Hanzi Machine Learning Existing continual learning literature relies heavily on a strong assumption that tasks arrive with a balanced data stream, which is often unrealistic in real-world applications. In this work, we explore task-imbalanced continual learning (TICL) scenarios where the distribution of task data is non-uniform across the whole learning process. We find that imbalanced tasks significantly challenge the capability of models to control the trade-off between stability and plasticity from the perspective of recent prompt-based continual learning methods. On top of the above finding, we propose Dynamically Anchored Prompting (DAP), a prompt-based method that only maintains a single general prompt to adapt to the shifts within a task stream dynamically. This general prompt is regularized in the prompt space with two specifically designed prompt anchors, called boosting anchor and stabilizing anchor, to balance stability and plasticity in TICL. Remarkably, DAP achieves this balance by only storing a prompt across the data stream, therefore offering a substantial advantage in rehearsal-free CL. Extensive experiments demonstrate that the proposed DAP results in 4.5% to 15% absolute improvements over state-of-the-art methods on benchmarks under task-imbalanced settings. Our code is available at https://github.com/chenxing6666/DAP |
| title | Dynamically Anchored Prompting for Task-Imbalanced Continual Learning |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2404.14721 |