Navigating Conflicting Views: Harnessing Trust for Learning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911017058959360 |
|---|---|
| author | Lu, Jueqing Buntine, Wray Qi, Yuanyuan Dipnall, Joanna Gabbe, Belinda Du, Lan |
| author_facet | Lu, Jueqing Buntine, Wray Qi, Yuanyuan Dipnall, Joanna Gabbe, Belinda Du, Lan |
| contents | Resolving conflicts is critical for improving the reliability of multi-view classification. While prior work focuses on learning consistent and informative representations across views, it often assumes perfect alignment and equal importance of all views, an assumption rarely met in real-world scenarios, as some views may express distinct information. To address this, we develop a computational trust-based discounting method that enhances the Evidential Multi-view framework by accounting for the instance-wise reliability of each view through a probability-sensitive trust mechanism. We evaluate our method on six real-world datasets using Top-1 Accuracy, Fleiss' Kappa, and a new metric, Multi-View Agreement with Ground Truth, to assess prediction reliability. We also assess the effectiveness of uncertainty in indicating prediction correctness via AUROC. Additionally, we test the scalability of our method through end-to-end training on a large-scale dataset. The experimental results show that computational trust can effectively resolve conflicts, paving the way for more reliable multi-view classification models in real-world applications. Codes available at: https://github.com/OverfitFlow/Trust4Conflict |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_00958 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Navigating Conflicting Views: Harnessing Trust for Learning Lu, Jueqing Buntine, Wray Qi, Yuanyuan Dipnall, Joanna Gabbe, Belinda Du, Lan Machine Learning Computer Vision and Pattern Recognition Resolving conflicts is critical for improving the reliability of multi-view classification. While prior work focuses on learning consistent and informative representations across views, it often assumes perfect alignment and equal importance of all views, an assumption rarely met in real-world scenarios, as some views may express distinct information. To address this, we develop a computational trust-based discounting method that enhances the Evidential Multi-view framework by accounting for the instance-wise reliability of each view through a probability-sensitive trust mechanism. We evaluate our method on six real-world datasets using Top-1 Accuracy, Fleiss' Kappa, and a new metric, Multi-View Agreement with Ground Truth, to assess prediction reliability. We also assess the effectiveness of uncertainty in indicating prediction correctness via AUROC. Additionally, we test the scalability of our method through end-to-end training on a large-scale dataset. The experimental results show that computational trust can effectively resolve conflicts, paving the way for more reliable multi-view classification models in real-world applications. Codes available at: https://github.com/OverfitFlow/Trust4Conflict |
| title | Navigating Conflicting Views: Harnessing Trust for Learning |
| topic | Machine Learning Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2406.00958 |