Responsible AI
Learn about responsible AI in Azure.
We'll cover the following...
Responsible AI
When we build models for the real world, it’s important to build responsible AI solutions. We need to make sure that our model is a fair representation of the real world.
There are multiple dimensions of responsible AI, including:
Fairness
Reliability and safety
Privacy and security
Inclusiveness
Transparency
Accountability
Azure Machine Learning can generate a responsible AI dashboard to do a deeper analysis of the model.
We will use the diabetes dataset here, which we previously used in the mini-project. For a simpler understanding, we've converted it to a classification problem: to predict if the person is diabetic or not. For simplicity and to balance the dataset, the dataset label is positive if their diabetic score is greater than 150.
You can see the sample train dataset below:
AGE SEX BMI BP S1 S2 S3 S4 S5 S6 Y59 2 32.1 101 157 93.2 38 4 4.8598 87 Diabetic48 1 21.6 87 183 103.2 70 3 3.8918 69 NonDiabetic72 2 30.5 93 156 93.6 41 4 4.6728 85 NonDiabetic