Research map / Label-efficient and robust learning / OOD and anomaly detection
Learning to Augment Distributions for Out-of-distribution Detection
Qizhou Wang, Zhen Fang, Yonggang Zhang, Feng Liu, Yixuan Li, Bo Han
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 |
|---|---|---|---|
| DAL (distributional augmentation) vs conventional outlier exposure | OE · CIFAR-10 average | FPR95: 4.67 -> 2.68 | Improved |
| DAL vs conventional outlier exposure | OE · CIFAR-10 average | AUROC: 98.88 -> 99.01 | Improved |
| DAL vs conventional outlier exposure | OE · CIFAR-100 average | FPR95: 43.14 -> 29.68 | Improved |
| DAL vs conventional outlier exposure | OE · CIFAR-100 average | AUROC: 90.27 -> 93.92 | Improved |
| DAL vs ATOM sampling strategy | ATOM · CIFAR-100 average | FPR95: 37.84 -> 29.68 | Improved |
| DAL vs POEM | POEM · CIFAR-100 average | FPR95: 34.77 -> 29.68 | Improved |
| DAL vs POEM | POEM · CIFAR-10 average | FPR95: 2.81 -> 2.68 | No clear effect |
| DAL vs DOE (worst-case OOD learning) | DOE · CIFAR-100 average | FPR95: 40.95 -> 29.68 | Improved |
| use auxiliary OOD data (OE) instead of ID data only | MSP · CIFAR-10 average | FPR95: 50.15 -> 4.67 | Improved |
| DAL vs OE on hard OOD | OE · CIFAR-10 vs CIFAR-100 | FPR95: 29.40 -> 25.45 | Improved |
| DAL vs POEM on LSUN-Fix | POEM · CIFAR-10 vs LSUN-Fix | FPR95: 1.24 -> 1.39 | Hurt |
| large Wasserstein radius rho=100 with large perturbation strength ps=10 (ablation) | DAL · CIFAR-100 | FPR95 | Hurt |
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.
- POEM — POEM: Out-of-Distribution Detection with Posterior Sampling (ICML 2022) · better on 6 of 7 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
- Image Background Serves as Good Proxy for Out-of-distribution Data (ICLR 2024 · similarity 0.91)
- Learning to Shape In-distribution Feature Space for Out-of-distribution Detection (NeurIPS 2024 · similarity 0.91)
- The Best of Both Worlds: On the Dilemma of Out-of-distribution Detection (NeurIPS 2024 · similarity 0.90)
- Can We Ignore Labels in Out of Distribution Detection? (ICLR 2025 · similarity 0.89)
- Fast Decision Boundary based Out-of-Distribution Detector (ICML 2024 · similarity 0.89)
- EAT: Towards Long-Tailed Out-of-Distribution Detection (AAAI 2024 · similarity 0.89)
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.
