Product DocumentationAPI and Python SDK ReferenceRelease Notes
Schedule a Demo
Schedule a Demo


Enrich your monitoring process with state-of-the-art techniques

Enrichments are additional services that the Arthur platform provides for state-of-the-art proactive model monitoring.

Enrichments in Arthur

  • Anomaly Detection: monitor and alert on incoming changes to your data distribution (compared to the reference dataset) based on complex interactions between features
  • Hotspots: automatically illuminate segments of underperformance within incoming inferences
  • Explainability: understand why your model is making decisions, by computing the importance of individual features from your data on your model's outcomes
  • Bias Mitigation: methods for model post-processing that improve the fairness of outcomes without re-deploying your model

Once activated, these enrichments are automatically computed on Arthur's backend, with results viewable in the online UI dashboard and queryable from Arthur's API.

Available Enrichments By Different Model Types

Due to the specialized nature of enrichments, they are only available for certain model types.

Model TypeAnomaly DetectionBias MitigationExplainabilityHot Spots
Tabular ClassificationXXXX
Tabular RegressionXX
Text ClassificationXX
Text RegressionXX
Text Sequence Generation (LLM)X (on inputs)
CV ClassificationXX
CV RegressionXX
CV Object DetectionX

Viewing Enabled Enrichments in the UI

You are also able to view the enrichments enabled for your specific model within the Arthur UI by clicking on the details sections of the model's overview page.

Enrichment Workflows

As enrichments are add-ons meant to enrich standard model monitoring, they run on their own workflows within Arthur.