Research map / Label-efficient and robust learning / OOD and anomaly detection
Memorizing Normality to Detect Anomaly: Memory-Augmented Deep Autoencoder for Unsupervised Anomaly Detection
Dong Gong, Lingqiao Liu, Vuong Le, Budhaditya Saha, Moussa Reda Mansour, Svetha Venkatesh, Anton van den Hengel
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 |
|---|---|---|---|
| add memory module to autoencoder (ablation) | AE · MNIST | average AUC: 0.9619 -> 0.9751 | Improved |
| add memory module to autoencoder (ablation) | AE · CIFAR-10 | average AUC: 0.5706 -> 0.6088 | Improved |
| add sparse shrinkage and entropy loss to memory addressing (ablation) | MemAE-nonSpar · MNIST | average AUC: 0.9725 -> 0.9751 | Improved |
| add sparse shrinkage and entropy loss to memory addressing (ablation) | MemAE-nonSpar · CIFAR-10 | average AUC: 0.6058 -> 0.6088 | Improved |
| add memory module to autoencoder (ablation) | AE · UCSD-Ped2 | AUC: 0.917 -> 0.941 | Improved |
| add memory module to autoencoder (ablation) | AE · CUHK Avenue | AUC: 0.810 -> 0.833 | Improved |
| add memory module to autoencoder (ablation) | AE · ShanghaiTech | AUC: 0.697 -> 0.712 | Improved |
| add sparse addressing to memory (ablation) | MemAE-nonSpar · ShanghaiTech | AUC: 0.688 -> 0.712 | Improved |
| add memory module to autoencoder (ablation) | AE · KDDCUP99 | F1: 0.9342 -> 0.9641 | Improved |
| add hard shrinkage to MemAE (ablation) | MemAE w/o Shrinkage · UCSD-Ped2 | AUC: 0.9324 -> 0.9410 | Improved |
| add entropy loss to MemAE (ablation) | MemAE w/o Entropy loss · UCSD-Ped2 | AUC: 0.9372 -> 0.9410 | Improved |
| add l1 sparsity regularization on latent features of AE (ablation) | AE · UCSD-Ped2 | AUC: 0.9170 -> 0.9286 | Improved |
Closest papers by meaning
- Learning Memory-Guided Normality for Anomaly Detection (CVPR 2020 · similarity 0.88)
- Two-Layer Convolutional Autoencoders Trained on Normal Data Provably Detect Unseen Anomalies (ICLR 2026 · similarity 0.88)
- Remembering Normality: Memory-guided Knowledge Distillation for Unsupervised Anomaly Detection (ICCV 2023 · similarity 0.88)
- A Comprehensive Augmentation Framework for Anomaly Detection (AAAI 2024 · similarity 0.86)
- Long-Tailed Anomaly Detection with Learnable Class Names (CVPR 2024 · similarity 0.85)
- ResAD: A Simple Framework for Class Generalizable Anomaly Detection (NeurIPS 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.
