Particle Competition and Cooperation for Robust Graph Convolutional Network Learning Under Label Noise
PCC+GCN refines noisy graph labels before training, slightly beating a baseline GCN with lower runtime.
PCC+GCN uses Particle Competition and Cooperation to detect suspicious labeled nodes and preserve, remove, or reassign their labels before Graph Convolutional Network training. The method was tested on ten NoisyGL datasets under Uniform, Pair, Random, and instance-dependent label noise, with hyperparameter analysis on Cora, CiteSeer, and PubMed. Under conventional noise it had the best average accuracy and rank, gaining 1.67 percentage points over a baseline GCN, and under instance-dependent noise it stayed competitive while being the fastest robust method on eight of ten datasets.
- PCC refines labels by preserving, removing, or reassigning suspicious nodes before GCN training.
- Tested on ten NoisyGL datasets under Uniform, Pair, Random, and instance-dependent noise.
- Average gain is 1.67 points over a baseline GCN across clean and conventional noise settings.
- Under instance-dependent noise it is the fastest robust method on eight of ten datasets.
Full article227 words · extracted from arxiv.org · click to collapse
Graph Convolutional Networks (GCNs) are highly sensitive to label noise, since corrupted supervision can propagate through the graph and degrade learned node representations. This work proposes PCC+GCN, a hybrid framework that uses Particle Competition and Cooperation (PCC) as a graph-based label-refinement stage before GCN training. PCC identifies suspicious labeled nodes through particle domination dynamics and determines whether their labels should be preserved, removed, or reassigned before GCN training. The framework also allows the graph used by PCC to be augmented with feature-based $k$-nearest-neighbor edges, while the GCN itself is trained on the original graph structure and node features. The proposed method was evaluated on ten graph datasets from the NoisyGL benchmark under conventional Uniform, Pair, and Random label noise, as well as under instance-dependent label noise. A detailed hyperparameter analysis was also conducted on Cora, CiteSeer, and PubMed. Under conventional noise, PCC+GCN achieved the highest overall average accuracy and the best average rank among the evaluated methods, with an average gain of $1.67$ percentage points over the baseline GCN across the clean setting and all noisy scenarios. Under instance-dependent noise, PCC+GCN remained competitive with the best-performing robust methods while requiring substantially lower execution time, being the fastest robust method on eight of the ten datasets. The results indicate that PCC-based label refinement provides an effective and computationally efficient preprocessing strategy for improving GCN robustness under noisy supervision.
Text extracted automatically; images, tables and formatting may be missing. Original: https://arxiv.org/abs/2609.22053