.png){kind=icon}
.png){kind=icon}
Return
Contrastive calibration on consensus and complementary multi-view representations
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2026.113291.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Proposes a self-representation NMF-based model for multi-view representation learning. • Combines model-level and late fusion to achieve comprehensive information integration. • Unifies consensus and complementary information via joint and disjoint factorizations. • Enhances representation discrimination through contrastive calibration regularization. • Demonstrates superiority over state-of-the-art methods on diverse multi-view datasets.
Keywords:
NMF-based multi-view learning
consensus and complementary information
joint and disjoint factorizations
contrastive calibration regularization
multi-view representation discrimination
Journal
IF:
7.6
Papers:
1.3W
Citations:
4.5W