Lifelong Learning of Graph Neural Networks for Open-World Node Classification
This work addresses the challenge of adapting GNNs to evolving real-world graphs with new classes, which is an incremental improvement for lifelong learning in graph-based AI applications.
The paper tackles the problem of lifelong learning for graph neural networks in open-world node classification, where graphs evolve and new classes emerge, by analyzing implicit and explicit knowledge retention. The results show that retaining only 50% of the GNN's receptive field maintains at least 95% accuracy compared to full history training, and implicit knowledge gains importance when explicit data is scarce.
Graph neural networks (GNNs) have emerged as the standard method for numerous tasks on graph-structured data such as node classification. However, real-world graphs are often evolving over time and even new classes may arise. We model these challenges as an instance of lifelong learning, in which a learner faces a sequence of tasks and may take over knowledge acquired in past tasks. Such knowledge may be stored explicitly as historic data or implicitly within model parameters. In this work, we systematically analyze the influence of implicit and explicit knowledge. Therefore, we present an incremental training method for lifelong learning on graphs and introduce a new measure based on $k$-neighborhood time differences to address variances in the historic data. We apply our training method to five representative GNN architectures and evaluate them on three new lifelong node classification datasets. Our results show that no more than 50% of the GNN's receptive field is necessary to retain at least 95% accuracy compared to training over the complete history of the graph data. Furthermore, our experiments confirm that implicit knowledge becomes more important when fewer explicit knowledge is available.