CertiPass LogoCertiPass

Creating Responsible AI Models with Amazon SageMaker

BEGINNER
60 minutes
5 tasks

In this lab, you will learn how to create AI models responsibly using Amazon SageMaker. You will explore how to utilize Amazon SageMaker Clarify to detect bias and ensure your models are transparent and explainable.

Scenario

You are a data scientist tasked with improving the fairness and transparency of machine learning models for your company. Your goal is to utilize Amazon SageMaker to build AI models that are responsible and align with industry standards for bias and fairness.

Learning Objectives

  • Understand features of responsible AI using SageMaker
  • Learn to use Amazon SageMaker Clarify for bias detection
  • Explore model explainability tools in SageMaker

tasks (5)

task 1: Set up an Amazon SageMaker notebook instance

10 min

task 2: Use Amazon SageMaker Clarify to detect bias in datasets

15 min

task 3: Implement model explainability using SageMaker tools

15 min

task 4: Analyze the tradeoffs between model safety and transparency

10 min

task 5: Apply human-centered design principles to explainable AI

15 min

Prerequisites

  • •Basic understanding of AI/ML concepts
  • •Familiarity with the AWS Management Console

Skills Tested

Understanding SageMaker for responsible AIUsing SageMaker Clarify for bias detectionApplying model explainability techniques

References

    Creating Responsible AI Models with Amazon SageMaker - Hands-On Lab - CertiPass