Research map / Probabilistic and scientific modeling / Variational and generative sampling
Deep Structured Energy Based Models for Anomaly Detection
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 |
|---|---|---|---|
| use energy score instead of reconstruction error as the anomaly criterion (ablation) | DSEBM · MNIST | F1: 0.9536 -> 0.9689 | Helped |
| use energy score instead of reconstruction error (ablation) | DSEBM · CIFAR-10 | F1: 0.8681 -> 0.8784 | Helped |
| use energy score instead of reconstruction error (ablation) | DSEBM · NATOPS | F1: 0.8792 -> 0.9022 | Helped |
| use energy score instead of reconstruction error (ablation) | DSEBM · FITNESS | F1: 0.8097 -> 0.8228 | Helped |
| use energy score instead of reconstruction error (ablation) | DSEBM · Caltech-101 | F1: 0.9077 -> 0.9159 | Helped |
| use energy score instead of reconstruction error (ablation) | DSEBM · Thyroid | F1: 0.8386 -> 0.8375 | Hurt |
| use energy score instead of reconstruction error (ablation) | DSEBM · KDD99 | F1: 0.7328 -> 0.7399 | Helped |
| use energy score instead of reconstruction error (ablation) | DSEBM · Usenet | F1: 0.7314 -> 0.7475 | Helped |
| DSEBM-e vs best baseline on high-dimensional Usenet (659 dims) | DSEBM-e · Usenet | recall: 0.7930 -> 0.8081 | Helped |
| DSEBM vs kernel methods on low-dimensional Thyroid (10 dims) | DSEBM · Thyroid | F1: 0.8481 -> 0.8375 | Hurt |
| add recurrent (RNN) energy function vs static baselines on sequential data | DSEBM-e · FITNESS | F1: 0.7501 -> 0.8228 | Helped |
| add recurrent energy function vs HMM | DSEBM-e · CUAVE | F1: 0.7507 -> 0.8402 | Helped |
Closest papers by meaning
- Energy-Based Models for Anomaly Detection: A Manifold Diffusion Recovery Approach (NeurIPS 2023 · similarity 0.88)
- Towards Training-free Anomaly Detection with Vision and Language Foundation Models (CVPR 2025 · similarity 0.83)
- Dense Projection for Anomaly Detection (AAAI 2024 · similarity 0.82)
- Deep Orthogonal Hypersphere Compression for Anomaly Detection (ICLR 2024 · similarity 0.82)
- We Need to Rethink Benchmarking in Anomaly Detection (NeurIPS 2026 · similarity 0.82)
- On Diffusion Modeling for Anomaly Detection (ICLR 2024 · similarity 0.82)
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.