ML Research Wiki / Benchmarks / Unsupervised Anomaly Detection / SMAP

SMAP

Unsupervised Anomaly Detection Benchmark

Performance Over Time

📊 Showing 7 results | 📏 Metric: F1

Top Performing Models

Rank Model Paper F1 Date Code
1 DFM (flow matching) DFM: Interpolant-free Dual Flow Matching 94.10 2024-10-11 -
2 ContextFlow++ (Glow-based) ContextFlow++: Generalist-Specialist Flow-based Generative Models with Mixed-Variable Context Encoding 93.62 2024-06-02 📦 gudovskiy/contextflow
3 TranAd TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Data 89.15 2022-01-18 📦 imperial-qore/tranad 📦 xuhongzuo/DeepOD
4 MTAD-GAT Multivariate Time-series Anomaly Detection via Graph Attention Network 88.80 2020-09-04 📦 ML4ITS/mtad-gat-pytorch 📦 mangushev/mtad-gat
5 CAE-M Unsupervised Deep Anomaly Detection for Multi-Sensor Time-Series Signals 88.27 2021-07-27 -
6 Glow Glow: Generative Flow with Invertible 1x1 Convolutions 86.05 2018-07-09 📦 openai/glow 📦 ikostrikov/pytorch-flows 📦 rosinality/glow-pytorch
7 GDN Graph Neural Network-Based Anomaly Detection in Multivariate Time Series 85.18 2021-06-13 📦 d-ailin/GDN 📦 huankoh/cst-gl

All Papers (7)