Research map / Label-efficient and robust learning / OOD and anomaly detection
Neural Transformation Learning for Deep Anomaly Detection Beyond Images
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 |
|---|---|---|---|
| Learn transformations jointly with an encoder using DCL | NTL · Time series one-vs-rest | AUC | Helped |
| Learn transformations jointly with an encoder using DCL | NTL · Epilepsy one-vs-rest | AUC: 82.6±1.7 -> 92.6±1.7 | Helped |
| Learn transformations instead of using hand-crafted transformations | Time series anomaly detection · RS one-vs-rest | AUC: 87.7±0.8 -> 86.5±0.6 | Hurt |
| Learn transformations jointly with an encoder using DCL | NTL · Time series n-vs-rest (n=N-1) | AUC | Mixed |
| Learn transformations jointly with an encoder using DCL | NTL · Tabular anomaly detection | F1-score (%) | Helped |
| Increase the number of learned transformations (ablation) | NTL · NATOPS and Arrhythmia | AUC / F1-score | Helped |
| Use K≤4 transformations rather than a larger set (ablation) | NTL · NATOPS and Arrhythmia | AUC / F1-score | Mixed |
| Vary transformation parameterization among feed-forward, residual, and multiplicative (ablation) | NTL · NATOPS and Arrhythmia | AUC / F1-score | No clear effect |
| Use the DCL objective to encourage semantic preservation and diversity | Neural transformation learning · Anomaly detection | Anomaly score | Helped |
Closest papers by meaning
- Language-Assisted Feature Transformation for Anomaly Detection (ICLR 2025 · similarity 0.87)
- Beyond Individual Input for Deep Anomaly Detection on Tabular Data (ICML 2024 · similarity 0.85)
- Self-supervised Transformation Learning for Equivariant Representations (NeurIPS 2024 · similarity 0.85)
- One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection (NeurIPS 2024 · similarity 0.85)
- A Diffusion-Based Framework for Multi-Class Anomaly Detection (AAAI 2024 · similarity 0.84)
- ResAD: A Simple Framework for Class Generalizable Anomaly Detection (NeurIPS 2024 · similarity 0.84)
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.