Stability and Generalization of lp-Regularized Stochastic Learning for GCN

2023/05/20 by Shiyu Liu, Linsen Wei, Liu, Shiyu +5
Computer Science · #Advanced Graph Neural Networks #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and ELM

paper · pdf · doi:10.48550/arxiv.2305.12085

openalex publication_date 2023/05/20 · openalex created_date 2023/05/27 · openalex updated_date 2026/07/28

Abstract

Graph convolutional networks (GCN) are viewed as one of the most popular representations among the variants of graph neural networks over graph data and have shown powerful performance in empirical experiments. That ℓ₂-based graph smoothing enforces the global smoothness of GCN, while (soft) ℓ₁-based sparse graph learning tends to promote signal sparsity to trade for discontinuity. This paper aims to quantify the trade-off of GCN between smoothness and sparsity, with the help of a general ℓ_p-regularized (1

Related

• Probabilistic and Regularized Graph Convolutional Networks
• Connecting Graph Convolutional Networks and Graph-Regularized PCA
• Stability and Generalization of Graph Convolutional Neural Networks
• Framework for Designing Filters of Spectral Graph Convolutional Neural Networks in the Context of Regularization Theory
• RandAlign: A Parameter-Free Method for Regularizing Graph Convolutional Networks