Open-source alternatives to NeuTraL AD

Revisiting VAE for Unsupervised Time Series Anomaly Detection: A Frequency Perspective •A new CVAE structure that using frequency as a condition. •Using global and local frequency information makes CVAE better reconstruct normal patterns.

KAN-AD: Time Series Anomaly Detection with Kolmogorov-Arnold Networks (ICML2025)

Code implementation for : Graph Neural Network-Based Anomaly Detection in Multivariate Time Series(AAAI'21)

Source code of CIKM'22 paper: TFAD: A Decomposition Time Series Anomaly Detection Architecture with Frequency Analysis Chaoli Zhang, Tian Zhou, Qingsong Wen, Liang Sun, "TFAD: A Decomposition Time Series Anomaly Detection Architecture with Time-Freq Analysis,” in Proc. 31st ACM International…

中文文档

This code is the official PyTorch implementation of our NeurIPS'25 Paper: CrossAD: Time Series Anomaly Detection with Cross-scale Associations and Cross-window Modeling.

Figure 1. Two different strategies for localizing temporal anomalies.

RANSynCoders (or RANCoders) is an unsupervised deep learning architecture for real-time anomaly detection and localizaiton within large multivariate time series. The method utilizes synchrony-analysis on latent representations for adjusting asynchronous variates fed into an encoder, bootstrap…

Offical implementation of "Graph-Augmented Normalizing Flows for Anomaly Detection of Multiple Time Series" (ICLR 2022). [paper](https://openreview.net/pdf?id=45L_dgP48Vd)

Exathlon

This is the implementation for the BeatGAN model architecture described in the paper: "BeatGAN: Anomalous Rhythm Detection using Adversarially Generated Time Series".

A curated list of awesome anomaly detection resources

This repository supplements our paper "TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Data" accepted in VLDB 2022. This is a refactored version of the code used for results in the paper for ease of use. Follow the below steps to replicate each cell in the…

ScatterAD leverages representation scattering as an inductive signal to jointly model temporal and topological patterns for effective multivariate time series anomaly detection.

An End-to-End Benchmark Suite for Univariate Time-Series Anomaly Detection

v2.0 updates: - Vectorized operations via numpy - Object-oriented restructure, improved organization - Merge branches into single branch for both processing modes (with/without labels) - Update requirements.txt and Dockerfile - Updated result output for both modes - PEP8 cleanup

ICML 2025 When Will It Fail? Anomaly to Prompt for Forecasting Future Anomalies in Time Series

Scripts and utility programs for implementing the USAD architecture.

Our implementation of MTAD-GAT: Multivariate Time-series Anomaly Detection (MTAD) via Graph Attention Networks (GAT) by Zhao et al. (2020).

The code of WWW (The Web Conference)'25 paper: "Can Multimodal LLMs Perform Time Series Anomaly Detection?"

This is the origin Pytorch implementation of CAT in the following paper: CAT: Beyond Efficient Transformer for Content-Aware AnomalyDetection in Event Sequences.

DONUT