Research map / Optimization / Optimizers
Mean-Variance Loss for Deep Age Estimation From a Face
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use the joint mean-variance and softmax losses instead of softmax loss alone (ablation) | AlexNet age estimation · MORPH II, RS | MAE (years): 3.324 -> 2.514 | Helped |
| Use the joint mean-variance and softmax losses instead of softmax loss alone (ablation) | AlexNet age estimation · MORPH II, SE | MAE (years): 4.043 -> 3.086 | Helped |
| Use the joint mean-variance and softmax losses instead of Euclidean loss alone (ablation) | AlexNet age estimation · MORPH II, RS | MAE (years): 3.289 -> 2.514 | Helped |
| Use the joint mean-variance and softmax losses instead of Euclidean loss alone (ablation) | AlexNet age estimation · MORPH II, SE | MAE (years): 3.932 -> 3.086 | Helped |
| Use mean-variance loss with softmax instead of a single-label regression or classification loss (ablation) | AlexNet age estimation · MORPH II, RS | CS (θ=5): 78.96% -> 88.90% | Helped |
| Use mean-variance loss with softmax instead of a single-label regression or classification loss (ablation) | AlexNet age estimation · MORPH II, SE | CS (θ=5): 73.24% -> 83.72% | Helped |
| Use mean-variance loss with softmax rather than DEX* | Age estimation · MORPH II, RS | MAE (years): 2.68 -> 2.16 | Helped |
| Use mean-variance loss with softmax rather than DIF | Age estimation · MORPH II, SE | MAE (years): 3.00 -> 2.79 | Helped |
| Use mean-variance loss with softmax rather than DEX* | Age estimation · FG-NET, LOPO | MAE (years): 3.09 -> 2.68 | Helped |
| Apply the age-estimation approach to image aesthetics assessment | AlexNet with mean-variance loss · AADB | ρ: 0.6782 -> 0.6647 | Hurt |
| Use a single proposed model rather than the OrangeLabs ensemble | Age estimation · CLAP2016 | ε-error: 0.2411 -> 0.2867 | Hurt |
Closest papers by meaning
- Deep Cost-Sensitive and Order-Preserving Feature Learning for Cross-Population Age Estimation (CVPR 2018 · similarity 0.88)
- Continuous Face Aging via Self-Estimated Residual Age Embedding (CVPR 2021 · similarity 0.87)
- PML: Progressive Margin Loss for Long-Tailed Age Classification (CVPR 2021 · similarity 0.87)
- Deep Regression Forests for Age Estimation (CVPR 2018 · similarity 0.87)
- From Cradle to Cane: A Two-Pass Framework for High-Fidelity Lifespan Face Aging (NeurIPS 2025 · similarity 0.84)
- A Call to Reflect on Evaluation Practices for Age Estimation: Comparative Analysis of the State-of-the-Art and a Unified Benchmark (CVPR 2024 · similarity 0.84)
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.