paper-with-me

홈 › Papers

Continual Few-shot Relation Learning via Embedding Space Regularization and Data Augmentation

2022-03-04 · ACL 2022 5 · Chengwei Qin, Shafiq Joty

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.

📄 PDF Abstract BibTeX arXiv:2203.02135

Code (1)

qcwthu/continual_fewshot_relation_learning 공식 구현 jax

Tasks

Data AugmentationRelation

Similar Papers 제목 키워드 기반

Continual Few-shot Relation Learning via Embedding Space Regularization and Data Augmentation

2021-11-16 · ACL ARR September 2021 9 · Anonymous

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 …

Data AugmentationRelation

Few-shot Continual Infomax Learning

2023-01-01 · ICCV 2023 1 · Ziqi Gu, Chunyan Xu, Jian Yang, Zhen Cui

Few-shot continual learning is the ability to continually train a neural network from a sequential stream of few-shot data. In this paper, we propose a Few-shot Continual Infomax Learning (FCIL) framework that makes …

Continual LearningFew-Shot Learning

Learning to Evolve: Bayesian-Guided Continual Knowledge Graph Embedding

2025-08-04 · Linyu Li, Zhi Jin, Yuanpeng He, Dongming Jin 외 arxiv

As social media and the World Wide Web become hubs for information dissemination, effectively organizing and understanding the vast amounts of dynamically evolving Web content is crucial. Knowledge graphs (KGs) provide a…

Knowledge Graph EmbeddingContinual LearningBayesian InferenceKnowledge Graphs

Lifelong Embedding Learning and Transfer for Growing Knowledge Graphs

2022-11-29 · Yuanning Cui, Yuxin Wang, Zequn Sun, Wenqiang Liu 외

Existing knowledge graph (KG) embedding models have primarily focused on static KGs. However, real-world KGs do not remain static, but rather evolve and grow in tandem with the development of KG applications. Consequentl…

Knowledge GraphsTransfer Learning

Elastic Weight Consolidation for Knowledge Graph Continual Learning: An Empirical Evaluation

2025-12-01 · Gaganpreet Jhajj, Fuhua Lin arxiv

Knowledge graphs (KGs) require continual updates as new information emerges, but neural embedding models suffer from catastrophic forgetting when learning new tasks sequentially. We evaluate Elastic Weight Consolidation …

Continual LearningKnowledge GraphsLink Prediction