Research map / Label-efficient and robust learning / Recognition and long-tail
RankMI: A Mutual Information Maximizing Ranking Loss
Check 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 RankMI instead of Margin loss | ResNet-50 image retrieval embeddings · CUB200-2011 | Recall@1: 63.6 -> 66.7 | Helped |
| Use RankMI instead of Margin loss | ResNet-50 image retrieval embeddings · CARS-196 | Recall@1: 79.6 -> 83.3; ABE-8: 85.2 | Mixed |
| Use RankMI instead of FastAP | ResNet-50 image retrieval embeddings · Stanford Online Products | Recall@1: 76.4 -> 74.3 | Hurt |
| Use RankMI instead of FastAP | ResNet-50 image retrieval embeddings · Stanford Online Products | Recall@1000: 98.2 -> 98.3 | Helped |
Closest papers by meaning
- RankDNN: Learning to Rank for Few-Shot Learning (AAAI 2023 · similarity 0.89)
- Fast Training of Triplet-Based Deep Binary Embedding Networks (CVPR 2016 · similarity 0.87)
- Ranked List Loss for Deep Metric Learning (CVPR 2019 · similarity 0.87)
- Supervised Metric Learning to Rank for Retrieval via Contextual Similarity Optimization (ICML 2023 · similarity 0.87)
- Exploiting Distribution Constraints for Scalable and Efficient Image Retrieval (ICLR 2025 · similarity 0.85)
- NUDGE: Lightweight Non-Parametric Fine-Tuning of Embeddings for Retrieval (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.