Research map / Label-efficient and robust learning / OOD and anomaly detection
Out of Distribution Data Detection Using Dropout Bayesian Neural Networks
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 |
|---|---|---|---|
| Add cosine randomized-embedding spread features to softmax-based Last features (ablation) | Dropout LeNet5 OOD detector · MNIST variants; OOD detection | AUC: 0.969 -> 0.979 | Helped |
| Add cosine randomized-embedding spread features to softmax-based Last features (ablation) | Dropout Char-CNN OOD detector · WiLI language OOD detection | AUC: 0.888 -> 0.926 | Helped |
| Add cosine randomized-embedding spread features to softmax-based Last features (ablation) | Dropout MalConv OOD detector · EMBER2018 malware OOD detection | AUC: 0.757 -> 0.791 | Helped |
| Add cosine randomized-embedding spread features to softmax-based Last features (ablation) | Dropout MalConv OOD detector · Brazilian malware OOD detection | AUC: 0.724 -> 0.839 | Helped |
| Add cosine randomized-embedding spread features to softmax-based Last features (ablation) | Dropout MalConv with LR OOD detector · EMBER2018 malware OOD detection, n=50/class | AUC: 0.786 -> 0.783 | Hurt |
| Add cosine randomized-embedding spread features to softmax-based Last features (ablation) | Dropout MalConv with LR OOD detector · Brazilian malware OOD detection, n=100/class | Recall: 0.645 -> 0.620 | Hurt |
| Use Euclidean distance instead of cosine distance for randomized embedding dispersion (ablation) | Dropout BNN OOD detection · MNIST vs. notMNIST | OOD separation | Hurt |
| Apply spectral normalization to the dropout LeNet5 (ablation) | MNIST-variant OOD detector · Image classification OOD detection | OOD detection performance | Helped |
Closest papers by meaning
- NECO: NEural Collapse Based Out-of-distribution detection (ICLR 2024 · similarity 0.86)
- Expecting The Unexpected: Towards Broad Out-Of-Distribution Detection (NeurIPS 2024 · similarity 0.86)
- The Best of Both Worlds: On the Dilemma of Out-of-distribution Detection (NeurIPS 2024 · similarity 0.85)
- DOS: Diverse Outlier Sampling for Out-of-Distribution Detection (ICLR 2024 · similarity 0.85)
- Mysteries of the Deep: Role of Intermediate Representations in Out of Distribution Detection (NeurIPS 2025 · similarity 0.85)
- EAT: Towards Long-Tailed Out-of-Distribution Detection (AAAI 2024 · 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.