Build Responsible AI using Error Analysis toolkit

Published: March 1, 2021, 11 p.m.

Error Analysis, a new Responsible AI open source toolkit, enables machine learning practitioners to identify model errors and diagnose the root causes behind these errors, helping to build responsible, reliable, and trusted solutions.

Jump to: 
[01:24] About AI models
[02:37] Error analysis 
[04:23] Interpretability and fairness 
[06:34] About Error Analysis toolkit  
[11:46] Error explorer demo 
[22:18] Debugging demos - Global and Local explanations

Learn more: 
Blog on MSFT Tech Community: https://aka.ms/AIShow/RAIBlog   
Website: https://aka.ms/AIShow/RAISite
Readme: https://aka.ms/AIShow/EAD
Sample notebook: https://aka.ms/AIShow/Notebooks/EA
Create a Free account (Azure)

Don't miss new episodes, subscribe to the AI Show join us every other Friday, for an AI Show livestream on Learn TV and YouTube https://aka.ms/LearnTV - https://aka.ms/AIShowLive