Analytic Drift Resister for Non-Exemplar Continual Graph Learning

arXiv cs.AI / 4/6/2026

💬 OpinionIdeas & Deep AnalysisModels & Research

共有:

Key Points

Non-Exemplar Continual Graph Learning (NECGL) avoids privacy risks by storing only class-level prototypes, but it introduces feature drift that can degrade continual learning performance.
The paper proposes Analytic Drift Resister (ADR), a theoretically grounded NECGL framework that uses iterative backpropagation to adapt beyond the limitations of frozen pre-trained models while improving model plasticity.
To counter drift caused by parameter updates, it introduces Hierarchical Analytic Merging (HAM), which performs layer-wise linear transformation merging in GNNs using ridge regression.
The framework further adds Analytic Classifier Reconstruction (ACR) to enable theoretically zero-forgetting class-incremental learning.
Experiments on four node classification benchmarks show ADR remains competitive with existing state-of-the-art approaches.

Abstract

Non-Exemplar Continual Graph Learning (NECGL) seeks to eliminate the privacy risks intrinsic to rehearsal-based paradigms by retaining solely class-level prototype representations rather than raw graph examples for mitigating catastrophic forgetting. However, this design choice inevitably precipitates feature drift. As a nascent alternative, Analytic Continual Learning (ACL) capitalizes on the intrinsic generalization properties of frozen pre-trained models to bolster continual learning performance. Nonetheless, a key drawback resides in the pronounced attenuation of model plasticity. To surmount these challenges, we propose Analytic Drift Resister (ADR), a novel and theoretically grounded NECGL framework. ADR exploits iterative backpropagation to break free from the frozen pre-trained constraint, adapting to evolving task graph distributions and fortifying model plasticity. Since parameter updates trigger feature drift, we further propose Hierarchical Analytic Merging (HAM), performing layer-wise merging of linear transformations in Graph Neural Networks (GNNs) via ridge regression, thereby ensuring absolute resistance to feature drift. On this basis, Analytic Classifier Reconstruction (ACR) enables theoretically zero-forgetting class-incremental learning. Empirical evaluation on four node classification benchmarks demonstrates that ADR maintains strong competitiveness against existing state-of-the-art methods.