Are Independently Estimated View Uncertainties Comparable? Unified Routing for Trusted Multi-View Classification

2026-04-10

AI Summary

This work addresses the challenge in multi-view classification that independently estimated view-specific uncertainties are incomparable due to the lack of cross-view consistency, causing fused predictions to be dominated by scale discrepancies among branches. To resolve this, the paper proposes a Trustworthy Multi-view Unified Routing framework (TMUR), which decouples view-specific evidence extraction from fusion arbitration and introduces a global context-aware unified router to generate sample-level expert weights for reliable uncertainty fusion. It is the first to reveal that independent evidence supervision fails to align evidence scales across views. The method further incorporates soft load balancing and diversity regularization to encourage expert specialization and balanced utilization. Theoretical analysis and experiments demonstrate that TMUR significantly outperforms local arbitration approaches in sample-dependent reliability scenarios, effectively enhancing both performance and trustworthiness in multi-view classification.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Reasoning under Uncertainty: Uncertainty Representations
• Computer Vision: Multi-modal Vision

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Trusted multi-view classification typically relies on a view-wise evidential fusion process: each view independently produces class evidence and uncertainty, and the final prediction is obtained by aggregating these independent opinions. While this design is modular and uncertainty-aware, it implicitly assumes that evidence from different views is numerically comparable. In practice, however, this assumption is fragile. Different views often differ in feature space, noise level, and semantic granularity, while independently trained branches are optimized only for prediction correctness, without any constraint enforcing cross-view consistency in evidence strength. As a result, the uncertainty used for fusion can be dominated by branch-specific scale bias rather than true sample-level reliability. To address this issue, we propose Trusted Multi-view learning with Unified Routing (TMUR), which decouples view-specific evidence extraction from fusion arbitration. TMUR uses view-private experts and one collaborative expert, and employs a unified router that observes the global multi-view context to generate sample-level expert weights. Soft load-balancing and diversity regularization further encourage balanced expert utilization and more discriminative expert specialization. We also provide theoretical analysis showing why independent evidential supervision does not identify a common cross-view evidence scale, and why unified global routing is preferable to branch-local arbitration when reliability is sample-dependent.

Problem

• multi-view classification
• evidential uncertainty
• cross-view comparability
• uncertainty calibration
• trusted fusion

Innovation

• multi-view classification
• evidential uncertainty
• unified routing
• expert specialization
• cross-view consistency