Feature-based Graph Attention Networks Improve Online Continual Learning

2025-02-13

AI Summary

To address catastrophic forgetting and dynamic data distribution adaptation in online continual image classification, this paper proposes a lightweight continual learning framework based on Graph Attention Networks (GATs). It innovatively transforms pretrained feature embeddings into multi-granularity dynamic graphs and designs a feature-driven GAT architecture. To preserve knowledge of past tasks under strict memory constraints, it introduces a rehearsal memory duplication mechanism and an enhanced hierarchical global pooling strategy. The method operates without task identifiers and supports single-pass streaming data. Evaluated on SVHN, CIFAR-10/100, and MiniImageNet, it consistently outperforms state-of-the-art approaches: average accuracy improves by 3.2–5.7 percentage points, and forgetting rate decreases by 41%, achieving a superior trade-off between stability and plasticity.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Online continual learning for image classification is crucial for models to adapt to new data while retaining knowledge of previously learned tasks. This capability is essential to address real-world challenges involving dynamic environments and evolving data distributions. Traditional approaches predominantly employ Convolutional Neural Networks, which are limited to processing images as grids and primarily capture local patterns rather than relational information. Although the emergence of transformer architectures has improved the ability to capture relationships, these models often require significantly larger resources. In this paper, we present a novel online continual learning framework based on Graph Attention Networks (GATs), which effectively capture contextual relationships and dynamically update the task-specific representation via learned attention weights. Our approach utilizes a pre-trained feature extractor to convert images into graphs using hierarchical feature maps, representing information at varying levels of granularity. These graphs are then processed by a GAT and incorporate an enhanced global pooling strategy to improve classification performance for continual learning. In addition, we propose the rehearsal memory duplication technique that improves the representation of the previous tasks while maintaining the memory budget. Comprehensive evaluations on benchmark datasets, including SVHN, CIFAR10, CIFAR100, and MiniImageNet, demonstrate the superiority of our method compared to the state-of-the-art methods.

Problem

• Online continual learning challenges
• Image classification adaptation
• Graph Attention Networks application

Innovation

• Graph Attention Networks
• Hierarchical feature maps
• Rehearsal memory duplication