SHAP

Description: Join Hayley to discuss how One AI makes interpreting models easy with robust reporting and the ability to model key findings and storyboards, in part, thanks to SHAP. SHAP is a framework for interpreting the output of machine learning models that clearly visualizes the importance of different features in model predictions.

Module Type: Foundational   & Functional

Level: Intermediate  

Audience: Model creators & managers; model storyboard viewers & stakeholders

Prerequisites: "One AI Recipes" module

Run time: ~11 minutes

Module Sections:

00:00 - Intro, Topics Covered, & Learning Outcomes

01:07 - Background & Overview of Shapley Values

03:30 - Intro to SHAP in Machine Learning

06:01 - Strengths & Weaknesses of SHAP

08:02 - Interpreting SHAP in One AI

10:38 - Conclusion & Thanks

The module transcription is attached below as a pdf.

• SHAP Transcript.docx.pdf80 KB