Abstract:An enhanced label propagation (LP) method called GraphHop was proposed recently. It outperforms graph convolutional networks (GCNs) in the semi-supervised node classification task on various networks. Although the performance of GraphHop was explained intuitively with joint node attribute and label signal smoothening, its rigorous mathematical treatment is lacking. In this paper, we propose a label efficient regularization and propagation (LERP) framework for graph node classification, and present an alternate optimization procedure for its solution. Furthermore, we show that GraphHop only offers an approximate solution to this framework and has two drawbacks. First, it includes all nodes in the classifier training without taking the reliability of pseudo-labeled nodes into account in the label update step. Second, it provides a rough approximation to the optimum of a subproblem in the label aggregation step. Based on the LERP framework, we propose a new method, named the LERP method, to solve these two shortcomings. LERP determines reliable pseudo-labels adaptively during the alternate optimization and provides a better approximation to the optimum with computational efficiency. Theoretical convergence of LERP is guaranteed. Extensive experiments are conducted to demonstrate the effectiveness and efficiency of LERP. That is, LERP outperforms all benchmarking methods, including GraphHop, consistently on five common test datasets, two large-scale networks, and an object recognition task at extremely low label rates (i.e., 1, 2, 4, 8, 16, and 20 labeled samples per class).

Label Propagation Through Adaptive Nearest Neighbor Based on Minimax Paths

A Bayesian Network nearest k-labels method for Multi-label classification

RW.KNN: a proposed random walk KNN algorithm for multi-label classification.

Label Distribution Learning-Enhanced Dual-KNN for Text Classification

Label Propagation Through Sparse Neighborhood and Its Applications

A Semi-supervised Kernel Learning Method Based on Label Propagation

Using Cluster Information to Improve Label Propagation

Neighbor selection for multilabel classification

A Label Propagation Approach Based on Local Optimization

A K-Nearest Neighbor Based Algorithm for Multi-Label Classification.

Transfer Learning for Nonparametric Classification: Minimax Rate and Adaptive Classifier

Label Propagation Through Linear Neighborhoods

Multi-Objective Optimization-Based Networked Multi-Label Active Learning.

Semi-supervised Classification Algorithm Using Fuzzy Nearest Neighborhood Label Propagation

Community Detection Using an Optimized Label Propagation Algorithm

ML-KNN: A lazy learning approach to multi-label learning

Contrastive Learning-Enhanced Nearest Neighbor Mechanism for Multi-Label Text Classification.

A label distance maximum-based classifier for multi-label learning

Learning K for Knn Classification

A Label Propagation Based User Locations Prediction Algorithm in Social Network

Label Efficient Regularization and Propagation for Graph Node Classification