Abstract:Graph-based semisupervised learning can explore the graph topology information behind the samples, becoming one of the most attractive research areas in machine learning in recent years. Nevertheless, existing graph-based methods also suffer from two shortcomings. On the one hand, the existing methods generate graphs in the original high-dimensional space, which are easily disturbed by noisy and redundancy features, resulting in low-quality constructed graphs that cannot accurately portray the relationships between data. On the other hand, most of the existing models are based on the Gaussian assumption, which cannot capture the local submanifold structure information of the data, thus reducing the discriminativeness of the learned low-dimensional representations. This article proposes a semisupervised subspace learning with adaptive pairwise graph embedding (APGE), which first builds a k<sub>1</sub> -nearest neighbor graph on the labeled data to learn local discriminant embeddings for exploring the intrinsic structure of the non-Gaussian labeled data, i.e., the submanifold structure. Then, a k<sub>2</sub> -nearest neighbor graph is constructed on all samples and mapped to GE learning to adaptively explore the global structure of all samples. Clustering unlabeled data and its corresponding labeled neighbors into the same submanifold, sharing the same label information, improves embedded data's discriminative ability. And the adaptive neighborhood learning method is used to learn the graph structure in the continuously optimized subspace to ensure that the optimal graph matrix and projection matrix are finally learned, which has strong robustness. Meanwhile, the rank constraint is added to the Laplacian matrix of the similarity matrix of all samples so that the connected components in the obtained similarity matrix are precisely equal to the number of classes in the sample, which makes the structure of the graph clearer and the relationship between the near-neighbor sample points more explicit. Finally, multiple experiments on several synthetic and real-world datasets show that the method performs well in exploring local structure and classification tasks.

Kernel Semi-Supervised Graph Embedding Model for Multimodal and Mixmodal Data.

Semi-Supervised Graph Based Embedding with Non-Convex Sparse Coding Techniques

Mutual Information-Based Graph Co-Attention Networks for Multimodal Prior-Guided Magnetic Resonance Imaging Segmentation

Semi-Supervised Graph Embedding Via Multi-instance Kernel Learning

Graph-based Multimodal Semi-Supervised Image Classification

Semisupervised Bayesian Gaussian Mixture Models for Non-Gaussian Soft Sensor

Multi-Modal Sentiment Classification With Independent and Interactive Knowledge via Semi-Supervised Learning

Semisupervised Subspace Learning With Adaptive Pairwise Graph Embedding

DisenSemi: Semi-supervised Graph Classification via Disentangled Representation Learning

Robust Semi-supervised Multimodal Medical Image Segmentation via Cross Modality Collaboration

$L_{\alpha}$-Regularization-Based Sparse Semi-Supervised Learning for Data with Complex Distributions

Multimodal graph learning based on 3D Haar semi-tight framelet for student engagement prediction

Semi-supervised Discriminant Analysis Based on Sparse-Coding Theory

Semi-Supervised Multiview Feature Selection With Adaptive Graph Learning

Comprehensive Semi-Supervised Multi-Modal Learning.

Semisupervised Feature Selection via Structured Manifold Learning

Self-paced Semi-Supervised Feature Selection with Application to Multi-Modal Alzheimer’s Disease Classification

Semi-Supervised Multi-Modal Learning with Incomplete Modalities

Flexible multi-view semi-supervised learning with unified graph

Flexible Multi-View Unsupervised Graph Embedding

Efficient Semi-Supervised Multimodal Hashing With Importance Differentiation Regression