Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Causal structure based root cause analysis of outliers

About

We describe a formal approach to identify 'root causes' of outliers observed in $n$ variables $X_1,\dots,X_n$ in a scenario where the causal relation between the variables is a known directed acyclic graph (DAG). To this end, we first introduce a systematic way to define outlier scores. Further, we introduce the concept of 'conditional outlier score' which measures whether a value of some variable is unexpected *given the value of its parents* in the DAG, if one were to assume that the causal structure and the corresponding conditional distributions are also valid for the anomaly. Finally, we quantify to what extent the high outlier score of some target variable can be attributed to outliers of its ancestors. This quantification is defined via Shapley values from cooperative game theory.

Dominik Janzing, Kailash Budhathoki, Lenon Minorics, Patrick Bl\"obaum• 2019

Related benchmarks

TaskDatasetResultRank
Root Cause AnalysisCausalMan
Top-1 Accuracy11
11
Root Cause AnalysisCausalChamber
Top-1 Accuracy26
11
Root Cause AnalysisER graph 50-node GT
Top-1 Accuracy30
11
Root Cause AnalysisER graph 50-node XGES-estimated
Top-1 Accuracy27
11
Root Cause AnalysisProRCA
Top-1 Accuracy50
11
Root Cause Analysis50-node ER graph (30% corruption)
Top-1 Accuracy18
11
Root Cause Analysis50-node ER graph 70% corruption
Top-1 Accuracy17
11
Root Cause Analysis50-node ER graph (50% corruption)
Top-1 Accuracy10
11
Root Cause AnalysisSynthetic Graph n=40
Top-5 Precision0.00e+0
11
Root Cause AnalysisSynthetic Graph n=60
Top-5 Precision0.00e+0
11
Showing 10 of 17 rows

Other info

Follow for update