Graph Lifelong Learning: A Survey

Abstract

Graph learning substantially contributes to solving artificial intelligence (AI) tasks in various graphrelated domains such as social networks, biological networks, recommender systems, and computer vision. However, despite its unprecedented prevalence, addressing the dynamic evolution of graph data over time remains a challenge. In many real-world applications, graph data continuously evolves. Current graph learning methods that assume graph representation is complete before the training process begins are not applicable in this setting. This challenge in graph learning motivates the development of a continuous learning process called graph lifelong learning to accommodate the future and refine the previous knowledge in graph data. Unlike existing survey papers that focus on either lifelong learning or graph learning separately, this survey paper covers the motivations, potentials, state-of-the-art approaches (that are well categorized), and open issues of graph lifelong learning. We expect extensive research and development interest in this emerging field.

Advancing autonomy through lifelong learning: a survey of autonomous intelligent systems

“…By doing so, it minimizes the risk of new task information significantly altering the previously acquired weights (Shaheen et al, 2022). An example of this approach is Elastic Weight Consolidation (EWC), which penalizes weight changes based on task importance, regularizing model parameters and preventing catastrophic forgetting of previous experiences (Febrinanto et al, 2022). • Rehearsal-based Approach: This method focuses on preserving knowledge by leveraging generative models to replay tasks whenever the model is modified or by storing samples from previously learned tasks in a memory buffer (Faber et al, 2023).…”