Research Article
Semi-supervised Learning via Adaptive Low-Rank Graph
@INPROCEEDINGS{10.1007/978-3-030-32388-2_37, author={Mingbo Zhao and Jiang Zhang and Cuili Yang}, title={Semi-supervised Learning via Adaptive Low-Rank Graph}, proceedings={Machine Learning and Intelligent Communications. 4th International Conference, MLICOM 2019, Nanjing, China, August 24--25, 2019, Proceedings}, proceedings_a={MLICOM}, year={2019}, month={10}, keywords={Semi-supervised learning Unsupervised learning Spectral clustering Adaptive low-rank model}, doi={10.1007/978-3-030-32388-2_37} }- Mingbo Zhao
Jiang Zhang
Cuili Yang
Year: 2019
Semi-supervised Learning via Adaptive Low-Rank Graph
MLICOM
Springer
DOI: 10.1007/978-3-030-32388-2_37
Abstract
Graph-based semi-supervised learning (SSL) is one of the most popular topics in the past decades. Most conventional graph-based SSL methods utilize two stage-approach to infer the class labels of the unlabeled data, i.e. it firstly constructs a graph for capturing the geometry of data manifold and then perform SSL for prediction. However, it suffers from three drawbacks: (1) the graph construction and SSL stages are separate. They do not share common information to enhance the performance of classification; (2) the graph construction and SSL should be scalable. However, most methods mainly focus on the improvement of classification accuracy but neglect the computational cost; (3) the graph should also be adaptive and robust to the parameters and datasets. However, this will usually increase computational cost making the efficiency cannot be guaranteed simultaneously. In this paper, we aim to handle the above issues. To achieve adaptiveness of SSL, we adopt a bilinear low-rank model for graph construction, where the coefficient matrix of the low-rank model is calculated through an adaptive and efficient procedure the corresponding constructed graph can capture the global structure of data manifold. Meriting from such a graph, we then propose a unified framework for scalable SSL, where we have involved the graph construction and SSL into a unified optimization problem. As a result, the discriminative information learned by SSL can be provided to improve the discriminative ability of graph construction, while the updated graph can further enhance the classification results of SSL. Simulation indicates that the proposed method can achieve better classification and clustering performance compared with other state-of-the-art graph-based SSL methods.