Continual Few-shot Relation Learning via Embedding Space Regularization and Data Augmentation
This addresses the challenge of expensive and time-consuming data labeling in real-world scenarios for continual relation learning, though it is incremental as it builds on existing methods with enhancements.
The paper tackles the problem of learning new relational patterns with very few labeled data while preventing catastrophic forgetting of previous tasks, achieving significant performance improvements over state-of-the-art methods in continual few-shot relation learning.
Existing continual relation learning (CRL) methods rely on plenty of labeled training data for learning a new task, which can be hard to acquire in real scenario as getting large and representative labeled data is often expensive and time-consuming. It is therefore necessary for the model to learn novel relational patterns with very few labeled data while avoiding catastrophic forgetting of previous task knowledge. In this paper, we formulate this challenging yet practical problem as continual few-shot relation learning (CFRL). Based on the finding that learning for new emerging few-shot tasks often results in feature distributions that are incompatible with previous tasks' learned distributions, we propose a novel method based on embedding space regularization and data augmentation. Our method generalizes to new few-shot tasks and avoids catastrophic forgetting of previous tasks by enforcing extra constraints on the relational embeddings and by adding extra {relevant} data in a self-supervised manner. With extensive experiments we demonstrate that our method can significantly outperform previous state-of-the-art methods in CFRL task settings.