Dynamically Anchored Prompting for Task-Imbalanced Continual Learning

• Main Authors: Hong, Chenxing, Jin, Yan, Kang, Zhiqi, Chen, Yizhou, Li, Mengke, Lu, Yang, Wang, Hanzi
• Format: Journal Article
• Language: English
• Published: 22.04.2024
• Subjects: Computer Science - 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-u...