Lightweight memory improves learning from few graph examples over time
A Lightweight Plastic-Memory Framework for Graph Few-Shot Class-Incremental Learning
Machine Learning
Summary
Learning to recognize new categories in graph data is hard when few examples exist, and learning many times can cause forgetting old knowledge. The authors created a small, adaptable memory module that updates dynamically with new information while remembering old categories. They combined this with a special training method that balances learning new tasks while keeping old skills stable. Their approach works better than older methods on several test datasets.
What this means in practice
- •For network engineers: Update network graph models efficiently with few labeled data from new devices without losing knowledge of existing devices.
- •For cybersecurity analysts: Maintain adaptive threat detection models on evolving graphs with minimal new training samples while preventing loss of previous threat patterns.
Authors
Zihan Mei, Zhili Qin, Tongze Zhang, Hongyuan Liu, Junming Shao, Qinli Yang
Abstract
Graph Incremental Learning has garnered increasing attention as dynamic graph data continues to emerge across diverse fields. Conventional approaches primarily address catastrophic forgetting by preserving node-related knowledge through replay or distillation techniques; however, they often incur high computational costs and inefficiency. This issue is further exacerbated in real-world scenarios where labeled data for new classes is scarce. In this paper, we propose a novel lightweight plastic-memory framework specifically designed for few-shot incremental learning on graphs. The core idea of our framework is the construction of a plastic-memory module that evolves over time, continuously updating and expanding its memory to accommodate new classes while retaining previously learned knowledge. In contrast to existing techniques, our memory module is both lightweight and effective, featuring an innovative evolving micro-clustering structure that dynamically updates representations of class prototypes, sub-prototypes, and their interaction weights. Building on this memory module, we introduce a memory-driven meta-learning framework that enhances adaptability to new tasks in its inner loop while maintaining stability for earlier tasks in the outer loop. Extensive experiments on four benchmark datasets demonstrate the framework's superior performance in balancing stability for old knowledge and adaptability to new knowledge.