Timezone: »

 
Poster
PIDForest: Anomaly Detection via Partial Identification
Parikshit Gopalan · Vatsal Sharan · Udi Wieder

Wed Dec 11 10:45 AM -- 12:45 PM (PST) @ East Exhibition Hall B + C #60
in Algorithms -- Unsupervised Learning »

We consider the problem of detecting anomalies in a large dataset. We propose a framework called Partial Identification which captures the intuition that anomalies are easy to distinguish from the overwhelming majority of points by relatively few attribute values. Formalizing this intuition, we propose a geometric anomaly measure for a point that we call PIDScore, which measures the minimum density of data points over all subcubes containing the point. We present PIDForest: a random forest based algorithm that finds anomalies based on this definition. We show that it performs favorably in comparison to several popular anomaly detection methods, across a broad range of benchmarks. PIDForest also provides a succinct explanation for why a point is labelled anomalous, by providing a set of features and ranges for them which are relatively uncommon in the dataset.

Keywords: Time Series Analysis · Unsupervised Learning · Algorithms · Algorithms -> Clustering; Algorithms -> Similarity and Distance Learning; Applications
[ Paper]  [ Poster]  [ Slides] 

Author Information

Parikshit Gopalan (VMware Research)
Vatsal Sharan (Stanford University)
Udi Wieder (VMware Research)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors