Regularized Contrastive Partial Multi-view Outlier Detection

Yijia Wang, Qianqian Xu, Yangbangyan Jiang, Siran Dai, Qingming Huang

Abstract: In recent years, multi-view outlier detection (MVOD) methods have advanced significantly, aiming to identify outliers within multi-view datasets. A key point is to better detect class outliers and class-attribute outliers, which only exist in multi-view data. However, existing methods either is not able to reduce the impact of outliers when learning view-consistent information, or struggle in cases with varying neighborhood structures. Moreover, most of them do not apply to partial multi-view data in real-world scenarios. To overcome these drawbacks, we propose a novel method named Regularized Contrastive Partial Multi-view Outlier Detection (RCPMOD). In this framework, we utilize contrastive learning to learn view-consistent information and distinguish outliers by the degree of consistency. Specifically, we propose (1) An outlier-aware contrastive loss with a potential outlier memory bank to eliminate their bias motivated by a theoretical analysis. (2) A neighbor alignment contrastive loss to capture the view-shared local structural correlation. (3) A spreading regularization loss to prevent the model from overfitting over outliers. With the Cross-view Relation Transfer technique, we could easily impute the missing view samples based on the features of neighbors. Experimental results on four benchmark datasets demonstrate that our proposed approach could outperform state-of-the-art competitors under different settings.

Primary Subject Area: [Content] Multimodal Fusion

Relevance To Conference: Multimedia data often exhibits multiple views, e.g., features sourced from various sensors or modalities, which provide complementary information about the same underlying entities or events. Consequently, multi-view learning plays a crucial role in improving the generalization performance in multimedia learning models. However, since the quality of data collection is difficult to control, outliers are inevitable in real-world multi-view datasets, and will severely harm multi-view learning. Meanwhile, certain views of some instances might be missing. Facing these challenges, this work focuses on detecting the outliers in partial multi-view datasets via multimodal/multi-view fusion. Experimental results show that the proposed framework greatly improves the detection performance, which will benefit other multi-view tasks in multimedia systems by preventing degradation caused by multi-view outliers.

Submission Number: 2680