Amazon SageMaker Clarify Bias Detection and Model Explainability

This topic describes how to configure an Amazon SageMaker Clarify processing job which can compute bias metrics and feature attributions for explainability. Clarify processing jobs are implemented using a specialized SageMaker Clarify container image. The following Topics list contains instructions on how to configure, run and troubleshoot a Clarify processing job and how to configure an analysis.

For additional information about bias metrics, explainability and how to interpret them, see Learn How Amazon SageMaker Clarify Helps Detect Bias, Fairness Measures for Machine Learning in Finance, and the Amazon AI Fairness and Explainability Whitepaper.

Topics

• How SageMaker Clarify Processing Jobs Work
• Get Started with a SageMaker Clarify Container
• Configure a SageMaker Clarify Processing Job
• Configure the Analysis
• Run SageMaker Clarify Processing Jobs for Bias Analysis and Explainability
• Detect Post-training Data and Model Bias with Amazon SageMaker Clarify
• Measure Post-training Data and Model Bias
• Amazon SageMaker Clarify Model Explainability
• Sample Notebooks
• Troubleshoot SageMaker Clarify Processing Jobs