Research map / Label-efficient and robust learning / Recognition and long-tail
Uncertainty Estimation for Multi-view Data: The Power of Seeing the Whole Picture
Myong Chol Jung, He Zhao, Joanna Dipnall, Belinda Gabbe, Lan Du
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| MGP vs TMC multi-view classifier | MGP · Handwritten, CUB, PIE, Caltech101, Scene15, HMDB | test accuracy: 98.10 -> 98.60 (Handwritten); 67.68 -> 70.00 (Scene15) | Improved |
| MGP vs DE (LF) in-domain | MGP · Handwritten, Scene15, PIE | test accuracy: 99.25 -> 98.60 (Handwritten); 67.05 -> 70.00 (Scene15) | Mixed |
| MGP vs TMC calibration | MGP · PIE | ECE: 0.072 -> 0.079 | Hurt |
| MGP vs DE (LF) calibration | MGP · Handwritten | ECE: 0.292 -> 0.006 | Improved |
| add Gaussian noise (std 0.01 to 10) to half of the views, MGP vs DE (LF) | MGP · Handwritten | average accuracy: 95.63 -> 97.66 | Improved |
| add Gaussian noise to half of the views, MGP vs DE (LF) | MGP · PIE | average accuracy: 67.69 -> 90.97 | Improved |
| larger noise range (std up to 10,000), MGP vs DE (LF) | MGP · HMDB | average accuracy: 57.10 -> 51.11 | Hurt |
| larger noise range (std up to 10,000), MGP vs TMC | MGP · PIE | average accuracy: 80.30 -> 90.15 | Improved |
| MGP vs DE (LF) on CIFAR10-C | MGP · CIFAR10-C | test accuracy: 75.40 -> 73.30 | Hurt |
| MGP vs DE (LF) on OOD detection SVHN | MGP · CIFAR10-C vs SVHN | OOD AUROC: 0.722 -> 0.803 | Improved |
| MGP vs TMC on OOD detection CIFAR100 | MGP · CIFAR10-C vs CIFAR100 | OOD AUROC: 0.675 -> 0.748 | Improved |
Closest papers by meaning
- Noisy Label Calibration for Multi-View Classification (AAAI 2025 · similarity 0.88)
- Deep Fuzzy Multi-view Learning for Reliable Classification (ICML 2025 · similarity 0.88)
- Enhancing Multi-View Classification Reliability with Adaptive Rejection (AAAI 2025 · similarity 0.88)
- Uncertainty-Aware Global-View Reconstruction for Multi-View Multi-Label Feature Selection (AAAI 2025 · similarity 0.87)
- SURE: SUrvey REcipes for building reliable and robust deep networks (CVPR 2024 · similarity 0.87)
- From Risk to Uncertainty: Generating Predictive Uncertainty Measures via Bayesian Estimation (ICLR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
