Research map / Graph learning / Graph self-supervised learning
Graph Regularized Autoencoder and its Application in Unsupervised Anomaly Detection
Imtiaz Ahmed, Travis Galoppo, Xia Hu, Yu Cheng Ding
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 MST-based embedding regularizer (LE) to autoencoder (ablation) | Autoencoder · 20 anomaly detection datasets | F1-score: 17 best results vs 0 for no regularizer | Improved |
| replace Euclidean-distance regularizer with MST regularizer (LE) | Autoencoder · Stamps | F1-score: 0.38 -> 0.69 | Improved |
| replace Euclidean-distance regularizer with MST regularizer (LE) | Autoencoder · Lymphography | F1-score: 0.67 -> 0.83 | Improved |
| replace GAE heat-kernel graph with MST regularizer (LE) | GAE · Parkinsons | F1-score: 0.20 -> 0.60 | Improved |
| use LE formulation instead of MDS formulation of MST regularizer (ablation) | MST regularizer · 20 datasets | uniquely best detections: 8 cases compared to 3 cases | Improved |
| use reconstruction loss instead of LoMST or COF detection on latent space (ablation) | MST (LE) autoencoder · KDD | F1-score: 0.02 -> 0.47 | Improved |
| use reconstruction loss instead of LoMST on latent space (ablation) | MST (LE) autoencoder · Ionosphere | F1-score: 0.71 vs 0.59 | Hurt |
| add traditional Gaussian noise denoising (ablation) | MST regularizer · 20 datasets | F1-score | No clear effect |
| replace regular denoising with neighbor-average denoising (ablation) | MST (LE) autoencoder · Parkinsons | F1-score: 0.60 -> 0.80 | Improved |
| add MST regularizer to AnoGAN | AnoGAN (1000 iterations) · KDD | F1-score: 0.44 -> 0.75 | Improved |
| add MST regularizer to AnoGAN | AnoGAN (100 iterations) · KDD | F1-score: 0.14 -> 0.35 | Improved |
| add MST regularizer to ALAD | ALAD (100 iterations) · KDD | F1-score: 0.04 -> 0.27 | Improved |
Closest papers by meaning
- ADA-GAD: Anomaly-Denoised Autoencoders for Graph Anomaly Detection (AAAI 2024 · similarity 0.86)
- Graph Geometry-Preserving Autoencoders (ICML 2024 · similarity 0.86)
- DiffGAD: A Diffusion-based Unsupervised Graph Anomaly Detector (ICLR 2025 · similarity 0.84)
- CurvGAD: Leveraging Curvature for Enhanced Graph Anomaly Detection (ICML 2025 · similarity 0.84)
- How Low Can You Go? Searching for the Intrinsic Dimensionality of Complex Networks using Metric Node Embeddings (ICLR 2025 · similarity 0.83)
- Generative Semi-supervised Graph Anomaly Detection (NeurIPS 2024 · similarity 0.83)
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.
