Adaptive Weighted LSSVM for Multi-View Classification

2025-12-02

AI Summary

Existing kernel-based multi-view methods suffer from the absence of explicit inter-view collaboration mechanisms and struggle to simultaneously achieve global coordination and privacy preservation. To address this, we propose Adaptive-Weighted Least Squares Support Vector Machines (AW-LSSVM). Our method introduces an iterative global coupling mechanism that dynamically guides each view to focus on samples misclassified by others, thereby enhancing inter-view complementarity; meanwhile, it preserves original feature isolation, inherently supporting privacy-sensitive scenarios. Its key innovation lies in replacing static fusion or fixed regularization with data-driven adaptive weight learning and coupling updates, jointly modeling both the strength and direction of inter-view collaboration. Extensive experiments on multiple benchmark datasets demonstrate that AW-LSSVM significantly outperforms state-of-the-art kernel-based multi-view approaches, achieving consistent improvements in both classification accuracy and privacy compatibility.

Technology Category

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

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Security and Privacy: Large-scale security measurements

Abstract

Multi-view learning integrates diverse representations of the same instances to improve performance. Most existing kernel-based multi-view learning methods use fusion techniques without enforcing an explicit collaboration type across views or co-regularization which limits global collaboration. We propose AW-LSSVM, an adaptive weighted LS-SVM that promotes complementary learning by an iterative global coupling to make each view focus on hard samples of others from previous iterations. Experiments demonstrate that AW-LSSVM outperforms existing kernel-based multi-view methods on most datasets, while keeping raw features isolated, making it also suitable for privacy-preserving scenarios.

Problem

• Enhances multi-view classification by adaptive weighted LS-SVM.
• Promotes complementary learning through iterative global coupling across views.
• Maintains feature isolation for privacy while improving performance.

Innovation

• Adaptive weighted LS-SVM for multi-view classification
• Iterative global coupling promotes complementary learning
• Keeps raw features isolated for privacy preservation