Research map / Optimization / Nonconvex and smooth optimization
Fast Stochastic AUC Maximization with $O(1/n)$-Convergence Rate
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 FSAUC (multi-stage primal-dual SG with adaptive radius/initialization) instead of SOLAM/OPAUC for one-pass AUC maximization | linear AUC maximization with square loss · eight LIBSVM benchmarks (Table 2) | final testing AUC | Helped |
| Use FSAUC instead of OPAUC on high-dimensional news20.binary | FSAUC vs OPAUC · news20.binary | final testing AUC: 0.975428 vs 0.967471 | Helped |
| Use FSAUC instead of OPAUC on w8a | FSAUC vs OPAUC · w8a | final testing AUC: 0.962703 vs 0.952536 (FSAUC higher) | Hurt |
| Use FSAUC instead of OPAUC on a9a | FSAUC vs OPAUC · a9a | final testing AUC: 0.900782 vs 0.900350 | Mixed |
| Use FSAUC instead of OPAUC on rcv1 binary | FSAUC vs OPAUC · rcv1 binary | final testing AUC: 0.995448 vs 0.990634 | Helped |
| Use FSAUC instead of SOLAM_l1 on rcv1 binary | FSAUC vs SOLAM_l1 · rcv1 binary | final testing AUC: 0.995448 vs 0.994767 | Helped |
Closest papers by meaning
- Stochastic AUC Maximization with Deep Neural Networks (ICLR 2020 · similarity 0.92)
- Stochastic Online AUC Maximization (NeurIPS 2016 · similarity 0.91)
- Stochastic Proximal Algorithms for AUC Maximization (ICML 2018 · similarity 0.91)
- Towards Nonlinear Sparse AUC Maximization via Compositional Stochastic Hard Thresholding (AAAI 2026 · similarity 0.91)
- High-probability complexity bounds for stochastic non-convex minimax optimization (NeurIPS 2024 · similarity 0.85)
- Stochastic Gradients under Nuisances (NeurIPS 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.