New method improves learning from incomplete multi-view multi-label data
When Semantically Consistent Encoding Meets View-Label Heterogeneity Modeling: A Unified Framework for Incomplete Multi-View Multi-Label Learning
Computer Vision and Pattern Recognition
Summary
When computers learn from data that has multiple types of information (views) and several labels, missing some views can make learning difficult. The authors created a new method called V2L that combines two ideas: making a shared understanding from incomplete views and deciding which views are most helpful for each label. Their approach lets the computer adapt to which pieces of information are most important for each prediction, even if some data is missing. Tests show this method performs well on several different tasks.
multi-view learningmulti-label classificationincomplete datasemantic representationvariational encodingfusion strategiesview-label relevancemachine learningadaptive weightinglatent representation
Authors
Chengliang Liu, Bo Li, Bob Zhang, Yanghao Zhou, Jie Wen, Wenwu Wang
Abstract
Incomplete multi-view multi-label learning requires not only robust semantic aggregation from partially observed views, but also label-aware exploitation of view-specific evidence. Existing approaches usually emphasize either shared representation learning or decision-level fusion. The former improves robustness against missing views, yet tends to compress label-discriminative view-specific cues into a single latent representation. The latter preserves individual view predictions, but often relies on fixed or globally learned fusion weights, ignoring that different labels of different instances may require different views. To address these limitations, this paper presents V2L, a unified representation-decision framework for incomplete multi-view multi-label classification. On the representation side, V2L constructs semantically consistent variational posteriors from incomplete views through a perturbation-aware encoding mechanism, which provides a stable shared semantic basis. On the decision side, V2L introduces an active view-label relevance modeling strategy that estimates instance-wise and label-wise view contributions, allowing each label prediction to adaptively select useful view-specific evidence. From the perspective of model architecture, these two important strategies are integrated into a unified framework through a hybrid fusion architecture, simultaneously meeting the requirements of cross-view semantic consistency and representational complementarity. Extensive experiments under both incomplete and complete settings show that V2L achieves leading performance on five benchmarks. Code is available at: https://github.com/justsmart/V2L.