Adaptive Collaborative Correlation Learning-based Semi-Supervised Multi-Label Feature Selection

TL;DR

This work tackles high-dimensional semi-supervised multi-label feature selection by introducing Access-MFS, a framework that simultaneously learns adaptive instance and label similarity graphs and selects discriminative yet uncorrelated features. It embeds feature selection into a generalized regression model with an extended uncorrelated constraint and a consistency term for labeled data, while coupling the regression with collaborative graph learning to reinforce the local structure of both feature and label spaces. An alternating optimization algorithm solves the joint objective, with convergence guarantees and a detailed complexity analysis. Empirical results on eight real-world datasets show Access-MFS consistently outperforms state-of-the-art methods, validating the benefit of jointly learning adaptive sample and label graphs for robust feature selection in semi-supervised multi-label settings.

Abstract

Semi-supervised multi-label feature selection has recently been developed to solve the curse of dimensionality problem in high-dimensional multi-label data with certain samples missing labels. Although many efforts have been made, most existing methods use a predefined graph approach to capture the sample similarity or the label correlation. In this manner, the presence of noise and outliers within the original feature space can undermine the reliability of the resulting sample similarity graph. It also fails to precisely depict the label correlation due to the existence of unknown labels. Besides, these methods only consider the discriminative power of selected features, while neglecting their redundancy. In this paper, we propose an Adaptive Collaborative Correlation lEarning-based Semi-Supervised Multi-label Feature Selection (Access-MFS) method to address these issues. Specifically, a generalized regression model equipped with an extended uncorrelated constraint is introduced to select discriminative yet irrelevant features and maintain consistency between predicted and ground-truth labels in labeled data, simultaneously. Then, the instance correlation and label correlation are integrated into the proposed regression model to adaptively learn both the sample similarity graph and the label similarity graph, which mutually enhance feature selection performance. Extensive experimental results demonstrate the superiority of the proposed Access-MFS over other state-of-the-art methods.

Figures (11)

• Figure 1: The framework of the proposed adaptive collaborative correlation learning-based semi-supervised multi-label feature selection method (Access-MFS).
• Figure 2: AP of different methods with varying numbers of selected features.
• Figure 3: MaF of different methods with varying numbers of selected features.
• Figure 4: RL of different methods with varying numbers of selected features.
• Figure 5: OE of different methods with varying numbers of selected features.

Theorems & Definitions (1)

• proof