Uncertainty Quantification for Incomplete Multi-View Data Using Divergence Measures

2025-07-14

AI Summary

To address inaccurate uncertainty quantification in incomplete multi-view data caused by noise and inter-modal discrepancies, this paper proposes KPHD-Net. The method replaces the KL divergence with the Hölder divergence for more robust measurement of categorical probability distribution discrepancies; models uncertainty via a variational Dirichlet distribution; and integrates Dempster–Shafer evidence theory with Kalman filtering to enable dynamic, robust fusion of multi-view information and future state prediction. Theoretical analysis guarantees consistency and convergence of the uncertainty estimation. Extensive experiments on multi-view classification and clustering tasks demonstrate that KPHD-Net significantly improves accuracy, noise robustness, and predictive reliability, comprehensively outperforming existing state-of-the-art approaches.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Intelligent Robots: State Estimation

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Existing multi-view classification and clustering methods typically improve task accuracy by leveraging and fusing information from different views. However, ensuring the reliability of multi-view integration and final decisions is crucial, particularly when dealing with noisy or corrupted data. Current methods often rely on Kullback-Leibler (KL) divergence to estimate uncertainty of network predictions, ignoring domain gaps between different modalities. To address this issue, KPHD-Net, based on Hölder divergence, is proposed for multi-view classification and clustering tasks. Generally, our KPHD-Net employs a variational Dirichlet distribution to represent class probability distributions, models evidences from different views, and then integrates it with Dempster-Shafer evidence theory (DST) to improve uncertainty estimation effects. Our theoretical analysis demonstrates that Proper Hölder divergence offers a more effective measure of distribution discrepancies, ensuring enhanced performance in multi-view learning. Moreover, Dempster-Shafer evidence theory, recognized for its superior performance in multi-view fusion tasks, is introduced and combined with the Kalman filter to provide future state estimations. This integration further enhances the reliability of the final fusion results. Extensive experiments show that the proposed KPHD-Net outperforms the current state-of-the-art methods in both classification and clustering tasks regarding accuracy, robustness, and reliability, with theoretical guarantees.

Problem

• Quantify uncertainty in incomplete multi-view data
• Improve reliability of multi-view integration with noisy data
• Address domain gaps in multi-view uncertainty estimation

Innovation

• Uses Hölder divergence for uncertainty quantification
• Integrates Dempster-Shafer theory with Kalman filter
• Employs variational Dirichlet for class distributions