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

TL;DR

This paper introduces KAFUSE, a novel multi-view unsupervised feature selection method that reduces redundancy and captures complex nonlinear dependencies by using kernel alignment and adaptive graph learning, improving data representation.

Contribution

The paper proposes a unified model combining kernel alignment with orthogonal constraints and sample-level adaptive graph fusion for enhanced multi-view feature selection.

Findings

KAFUSE outperforms state-of-the-art methods on real datasets.

It effectively captures nonlinear feature dependencies.

Adaptive graph learning improves local structure preservation.

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.