Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Probabilistic Deep Ordinal Regression Based on Gaussian Processes
Yanzhu Liu, Fan Wang, Adams Wai Kin Kong
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 |
|---|---|---|---|
| train CNN and GP jointly end-to-end instead of separately (logistic likelihood) (ablation) | GP-DNNOR · Urban image aesthetics | accuracy (%): 67.37 -> 68.29 | Improved |
| train CNN and GP jointly end-to-end instead of separately (logistic likelihood) (ablation) | GP-DNNOR · Urban image aesthetics | MAE: 0.34 -> 0.32 | Improved |
| use logistic instead of Gaussian ordinal likelihood (joint training) (ablation) | GP-DNNOR · Urban image aesthetics | accuracy (%): 67.32 -> 68.29 | Improved |
| use logistic instead of Gaussian ordinal likelihood (joint training) (ablation) | GP-DNNOR · Urban image aesthetics | MAE: 0.34 -> 0.32 | Improved |
| use logistic instead of Gaussian likelihood when training separately (ablation) | CNN + GP · Urban image aesthetics | MAE: 0.35 -> 0.34 | No clear effect |
| replace multi-class GP layer (CNNm-GP) with ordinal GP layer with learned boundaries | GP-DNNOR · Urban image aesthetics | MAE: 0.41 -> 0.32 | Improved |
| replace multi-class GP layer with ordinal GP layer | GP-DNNOR · Adience Face | accuracy (%): 46.1 -> 57.4 | Improved |
| replace multi-class GP layer with ordinal GP layer | GP-DNNOR · Historical Color Image | accuracy (%): 49.98 -> 46.60 | Hurt |
| replace multi-class GP layer with ordinal GP layer | GP-DNNOR · Historical Color Image | MAE: 0.84 -> 0.76 | Improved |
| replace softmax classifier CNNm with ordinal GP layer | GP-DNNOR · Adience Face | MAE: 0.61 -> 0.54 | Improved |
| compare GP-DNNOR with CNNPOR on accuracy | GP-DNNOR · Historical Color Image | accuracy (%): 50.12 -> 46.60 | Hurt |
Closest papers by meaning
- Soft Labels for Ordinal Regression (CVPR 2019 · similarity 0.86)
- A Constrained Deep Neural Network for Ordinal Regression (CVPR 2018 · similarity 0.85)
- Learning Probabilistic Ordinal Embeddings for Uncertainty-Aware Regression (CVPR 2021 · similarity 0.85)
- Deep Gaussian Processes for Regression using Approximate Expectation Propagation (ICML 2016 · similarity 0.85)
- SLACE: A Monotone and Balance-Sensitive Loss Function for Ordinal Regression (AAAI 2025 · similarity 0.83)
- Neural Collapse in Cumulative Link Models for Ordinal Regression: An Analysis with Unconstrained Feature Model (NeurIPS 2025 · similarity 0.82)
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.
