juhar1github/Awesome-causal-time-series-anomaly-detection

A curated collection of causality-based time-series anomaly detection and root-cause analysis papers with official code, datasets, and benchmarks

Python

0

12 commits

updated Jul 29, 2026

See the code

README

Awesome Causal Time-Series Anomaly Detection

A curated collection of papers, official implementations, libraries, datasets, and benchmarks for causality-based time-series anomaly detection and root-cause analysis.

Scope

This collection includes work where causality is central to at least one of the following:

  1. Anomaly scoring or detection
  2. Granger-causal or structural-causal graph discovery
  3. Root-cause localization or intervention recognition
  4. Causality-aware representation learning
  5. Causal benchmarks for multivariate time series

The main list prioritizes official author repositories. A blank code cell means that no verified public implementation was located.

Papers

YearMethodPaperVenuePaper linkCode linkMain causal idea
2026CGTCausally-Constrained Probabilistic Forecasting for Time-Series Anomaly DetectionarXivPaperCodeTime-lagged causal graph masks a probabilistic forecasting model
2026CAADCAAD: Causality-Aware Multivariate Time Series Anomaly Detection via Multi-Scale Alignment and Structural Causal ConsistencyarXivPaperContinuous verification of Granger-causal consistency
2025AERCARoot Cause Analysis of Anomalies in Multivariate Time Series through Granger Causal DiscoveryICLRPaperCodeJoint Granger causal discovery and intervention-based root-cause scoring
2025CAROTSCausality-Aware Contrastive Learning for Robust Multivariate Time-Series Anomaly DetectionICMLPaperCodeCausality-aware contrastive representation learning
2025causRCAcausRCA Dataset and Benchmarking FrameworkDataset and benchmarkNot availableCodeGround-truth causal structures and fault scenarios
2025CGADEntropy Causal Graphs for Multivariate Time Series Anomaly DetectionACM TISTPaperCodeTransfer-entropy causal graph with graph and temporal convolutions
2025RCAEvalRCAEval: A Benchmark for Root Cause Analysis of Microservice Systems with Telemetry DataThe Web ConferencePaperCodeIncludes reproducible causal and graph-based RCA baselines
2025GCADGCAD: Anomaly Detection in Multivariate Time Series from the Perspective of Granger CausalityarXivPaperDynamic gradient-based Granger-causality discovery
2025OracleADStructured Temporal Causality for Interpretable Multivariate Time-Series Anomaly DetectionarXivPaperStructured temporal causal modeling
2024RUNRoot Cause Analysis in Microservice Using Neural Granger Causal DiscoveryAAAIPaperCodeContrastive temporal encoder plus neural Granger discovery and personalized PageRank
2023EasyRCARoot Cause Identification for Collective Anomalies in Time Series given an Acyclic Summary Causal Graph with LoopsAISTATSPaperCodeCausal graph, anomaly timing, d-separation, and direct-effect changes
2023PyRCAPyRCA: A Library for Metric-based Root Cause AnalysisarXiv / open-source libraryPaperCodeUnified causal-graph construction and graph-based RCA methods
2023CausalADA Causal Approach to Detecting Multivariate Time-Series Anomalies and Root CausesICLRPaperDetects violations of local causal mechanisms
2022CIRCACausal Inference-Based Root Cause Analysis for Online Service Systems with Intervention RecognitionKDDPaperCodeRecognizes interventions through changes in conditional mechanisms

This repository links to external implementations. It does not copy or redistribute another author's code.

Before reusing an implementation:

  1. Read the original repository's license
  2. Preserve required copyright and license notices
  3. Cite the associated paper
  4. Mark unofficial implementations clearly
  5. Do not redistribute code that has no reuse license

Search keywords

causal time-series anomaly detection, multivariate time-series anomaly detection, Granger causality, causal discovery, structural causal model, root-cause analysis, intervention recognition, causal graph, AIOps, industrial monitoring

Disclaimer

Metadata can change when preprints are accepted or repositories are released. Verify paper versions, venues, and licenses before reuse.

Contributors

juhar1github

12 commits

juhar1github/Awesome-causal-time-series-anomaly-detection

A curated collection of causality-based time-series anomaly detection and root-cause analysis papers with official code, datasets, and benchmarks

Python

0

12 commits

updated Jul 29, 2026

See the code

README

Awesome Causal Time-Series Anomaly Detection

A curated collection of papers, official implementations, libraries, datasets, and benchmarks for causality-based time-series anomaly detection and root-cause analysis.

Scope

This collection includes work where causality is central to at least one of the following:

  1. Anomaly scoring or detection
  2. Granger-causal or structural-causal graph discovery
  3. Root-cause localization or intervention recognition
  4. Causality-aware representation learning
  5. Causal benchmarks for multivariate time series

The main list prioritizes official author repositories. A blank code cell means that no verified public implementation was located.

Papers

YearMethodPaperVenuePaper linkCode linkMain causal idea
2026CGTCausally-Constrained Probabilistic Forecasting for Time-Series Anomaly DetectionarXivPaperCodeTime-lagged causal graph masks a probabilistic forecasting model
2026CAADCAAD: Causality-Aware Multivariate Time Series Anomaly Detection via Multi-Scale Alignment and Structural Causal ConsistencyarXivPaperContinuous verification of Granger-causal consistency
2025AERCARoot Cause Analysis of Anomalies in Multivariate Time Series through Granger Causal DiscoveryICLRPaperCodeJoint Granger causal discovery and intervention-based root-cause scoring
2025CAROTSCausality-Aware Contrastive Learning for Robust Multivariate Time-Series Anomaly DetectionICMLPaperCodeCausality-aware contrastive representation learning
2025causRCAcausRCA Dataset and Benchmarking FrameworkDataset and benchmarkNot availableCodeGround-truth causal structures and fault scenarios
2025CGADEntropy Causal Graphs for Multivariate Time Series Anomaly DetectionACM TISTPaperCodeTransfer-entropy causal graph with graph and temporal convolutions
2025RCAEvalRCAEval: A Benchmark for Root Cause Analysis of Microservice Systems with Telemetry DataThe Web ConferencePaperCodeIncludes reproducible causal and graph-based RCA baselines
2025GCADGCAD: Anomaly Detection in Multivariate Time Series from the Perspective of Granger CausalityarXivPaperDynamic gradient-based Granger-causality discovery
2025OracleADStructured Temporal Causality for Interpretable Multivariate Time-Series Anomaly DetectionarXivPaperStructured temporal causal modeling
2024RUNRoot Cause Analysis in Microservice Using Neural Granger Causal DiscoveryAAAIPaperCodeContrastive temporal encoder plus neural Granger discovery and personalized PageRank
2023EasyRCARoot Cause Identification for Collective Anomalies in Time Series given an Acyclic Summary Causal Graph with LoopsAISTATSPaperCodeCausal graph, anomaly timing, d-separation, and direct-effect changes
2023PyRCAPyRCA: A Library for Metric-based Root Cause AnalysisarXiv / open-source libraryPaperCodeUnified causal-graph construction and graph-based RCA methods
2023CausalADA Causal Approach to Detecting Multivariate Time-Series Anomalies and Root CausesICLRPaperDetects violations of local causal mechanisms
2022CIRCACausal Inference-Based Root Cause Analysis for Online Service Systems with Intervention RecognitionKDDPaperCodeRecognizes interventions through changes in conditional mechanisms

This repository links to external implementations. It does not copy or redistribute another author's code.

Before reusing an implementation:

  1. Read the original repository's license
  2. Preserve required copyright and license notices
  3. Cite the associated paper
  4. Mark unofficial implementations clearly
  5. Do not redistribute code that has no reuse license

Search keywords

causal time-series anomaly detection, multivariate time-series anomaly detection, Granger causality, causal discovery, structural causal model, root-cause analysis, intervention recognition, causal graph, AIOps, industrial monitoring

Disclaimer

Metadata can change when preprints are accepted or repositories are released. Verify paper versions, venues, and licenses before reuse.

Contributors

juhar1github

12 commits

Languages

Python

100.0%