Detect Bias & Ensure Fairness in Domino
This session will utilize the IBM AI Fairness 360 library to detect bias in data and models, and produce a "fair" model in Domino.
Using open-source libraries to detect bias and ensure fairness with Domino
By the end of this customer tech hour, you will be able to:
- Recognize many ways of measuring “bias” and “fairness” in data science.
- Identify bias/fairness metrics that make sense for your context.
- Detect bias in data and in models with the IBM AI Fairness 360 library.
- Produce a “fair” model using the IBM AI Fairness 360 library.
- Illustrate how Domino’s governance features make Responsible AI, detecting bias, and ensuring fairness easier.