Research map / Label-efficient and robust learning / Recognition and long-tail
Correcting the Triplet Selection Bias for Triplet Loss
Baosheng Yu, Tongliang Liu, Mingming Gong, Changxing Ding, Dacheng Tao
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 |
|---|---|---|---|
| add distribution matching loss (lambda=0.01) | Triplet loss · CARS196 | R@1: 0.7781 -> 0.7917 | Improved |
| add distribution matching loss (lambda=0.005) | Triplet loss · CARS196 | R@1: 0.7781 -> 0.7912 | Improved |
| increase lambda to 0.1 | Triplet loss · CARS196 | R@1: 0.7917 -> 0.7631 | Hurt |
| add distribution matching loss (lambda=0.01) | Triplet loss · CUB200-2011 | R@1: 0.4450 -> 0.4660 | Improved |
| add distribution matching loss (lambda=0.005) | Triplet loss · CUB200-2011 | R@1: 0.4450 -> 0.4439 | No clear effect |
| increase lambda to 0.5 | Triplet loss · CUB200-2011 | R@1: 0.4660 -> 0.4483 | Hurt |
| add distribution matching loss (lambda=0.01) | Triplet loss · Stanford Online Products | R@1: 0.6274 -> 0.6303 | No clear effect |
| add distribution matching loss (lambda=0.01) | Triplet loss · Stanford Online Products | R@100: 0.9050 -> 0.9071 | No clear effect |
| add distribution matching loss (lambda=0.01) | Triplet loss · CARS196 | R@20: 0.9716 -> 0.9745 | Improved |
| add distribution matching loss (lambda=0.01) | Triplet loss · CUB200-2011 | R@20: 0.8893 -> 0.9024 | Improved |
| add distribution matching loss to triplet loss | Triplet loss · MNIST | test accuracy | Improved |
| vary lambda over 0, 0.1, 0.5, 1.0, 2.0, 5.0 | Triplet loss · MNIST | test accuracy | Mixed |
Closest papers by meaning
- Deep Metric Learning with Hierarchical Triplet Loss (ECCV 2018 · similarity 0.91)
- Hard negative examples are hard, but useful (ECCV 2020 · similarity 0.90)
- On the Stability and Generalization of Triplet Learning (AAAI 2023 · similarity 0.88)
- Triplet-Center Loss for Multi-View 3D Object Retrieval (CVPR 2018 · similarity 0.88)
- Mathematical Justification of Hard Negative Mining via Isometric Approximation Theorem (ICLR 2024 · similarity 0.85)
- Learning with Noisy Triplet Correspondence for Composed Image Retrieval (CVPR 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.
