Research map / Optimization / Optimizers
Exploring the Algorithm-Dependent Generalization of AUPRC Optimization with List Stability
Peisong Wen, Qianqian Xu, Zhiyong Yang, Yuan He, Qingming Huang
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 |
|---|---|---|---|
| use unbiased stochastic estimator (prior pi) instead of biased np/(np+nn) (ablation) | SOPRC with SGD · iNaturalist | mAUPRC: 34.58 -> 35.84 | Improved |
| add auxiliary vector v_t to estimate positive scores (ablation) | SOPRC with unbiased estimator · iNaturalist | mAUPRC: 35.84 -> 35.99 | Improved |
| add semi-variance regularization L_var (ablation) | SOPRC with v_t · iNaturalist | mAUPRC: 35.99 -> 36.16 | Improved |
| switch optimizer from SGD to Adam (ablation) | SOPRC full · iNaturalist | mAUPRC: 36.16 -> 36.20 | No clear effect |
| add unbiased estimator, v_t and L_var together (ablation) | SOPRC · iNaturalist | Recall@4: 66.35 -> 68.22 | Improved |
| add unbiased estimator, v_t and L_var together (ablation) | SOPRC · iNaturalist | Recall@32: 92.72 -> 93.71 | Improved |
| use larger batch size | SOPRC · iNaturalist validation | mAUPRC | Improved |
| optimize AUPRC directly instead of pairwise losses (XBM) | SOPRC vs XBM · iNaturalist | mAUPRC: 27.46 -> 36.16 | Improved |
| optimize AUPRC directly instead of pairwise losses (XBM) | SOPRC vs XBM · PKU VehicleID | Recall@1: 92.78 -> 92.56 | Hurt |
| SOPRC versus best baseline SmoothAP | SOPRC · Stanford Online Products | mAUPRC: 61.65 -> 62.75 | Improved |
| SOPRC versus best baseline SmoothAP | SOPRC · iNaturalist | mAUPRC: 33.92 -> 36.16 | Improved |
| SOPRC versus best baseline DIR | SOPRC · PKU VehicleID | mAUPRC: 72.72 -> 74.92 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SmoothAP — Smooth-AP: Smoothing the Path Towards Large-Scale Image Retrieval (ECCV 2020) · better on 9 of 9 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Algorithm-Dependent Generalization of AUPRC Optimization: Theory and Algorithm (TPAMI 2024 · similarity 0.97)
- Stochastic Optimization of Areas Under Precision-Recall Curves with Provable Convergence (NeurIPS 2021 · similarity 0.89)
- Fast Stochastic AUC Maximization with $O(1/n)$-Convergence Rate (ICML 2018 · similarity 0.89)
- Algorithmic Stability Unleashed: Generalization Bounds with Unbounded Losses (ICML 2024 · similarity 0.87)
- Stability and Generalization of Stochastic Compositional Gradient Descent Algorithms (ICML 2024 · similarity 0.86)
- Generalization Analysis of Biased Stochastic Gradient Methods for Minimax Problems (NeurIPS 2026 · similarity 0.86)
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.
