Kernel Alignment-based Multi-view Unsupervised Feature Selection with Sample-level Adaptive Graph Learning

2026-01-12

AI Summary

Existing unsupervised multi-view feature selection methods struggle to effectively model nonlinear feature dependencies and typically employ sample-invariant graph fusion strategies that overlook the heterogeneity of local neighborhood structures. To address these limitations, this work proposes a unified framework that jointly captures both linear and nonlinear feature redundancies through kernel alignment and orthogonality constraints. Furthermore, it introduces a sample-level adaptive graph fusion mechanism that dynamically adjusts the contribution weight of each view for every individual sample, enabling precise integration of multi-view similarity graphs. By mutually reinforcing feature selection and graph learning within a single optimization process, the proposed method consistently outperforms state-of-the-art approaches across multiple real-world multi-view datasets, demonstrating its effectiveness and superiority.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Although multi-view unsupervised feature selection (MUFS) has demonstrated success in dimensionality reduction for unlabeled multi-view data, most existing methods reduce feature redundancy by focusing on linear correlations among features but often overlook complex nonlinear dependencies. This limits the effectiveness of feature selection. In addition, existing methods fuse similarity graphs from multiple views by employing sample-invariant weights to preserve local structure. However, this process fails to account for differences in local neighborhood clarity among samples within each view, thereby hindering accurate characterization of the intrinsic local structure of the data. In this paper, we propose a Kernel Alignment-based multi-view unsupervised FeatUre selection with Sample-level adaptive graph lEarning method (KAFUSE) to address these issues. Specifically, we first employ kernel alignment with an orthogonal constraint to reduce feature redundancy in both linear and nonlinear relationships. Then, a cross-view consistent similarity graph is learned by applying sample-level fusion to each slice of a tensor formed by stacking similarity graphs from different views, which automatically adjusts the view weights for each sample during fusion. These two steps are integrated into a unified model for feature selection, enabling mutual enhancement between them. Extensive experiments on real multi-view datasets demonstrate the superiority of KAFUSE over state-of-the-art methods.

Problem

• multi-view unsupervised feature selection
• nonlinear dependencies
• similarity graph fusion
• sample-level adaptive weighting
• local structure preservation

Innovation

• Kernel Alignment
• Sample-level Adaptive Graph Learning
• Multi-view Unsupervised Feature Selection
• Nonlinear Dependency Modeling
• Cross-view Consistent Graph