What approaches have you used to reduce bias and disparate impact in model selection?
The Genesys approach to managing AI bias and model impact is focused primarily on:
- Assessment and curation of model input data to make certain that no sensitive features are included in the models.
- Model monitoring tracks various model metrics to help avoid data drift and concept drift.
- Model cards and dataset cards document various characteristics of AI models and training datasets in a standardized format. These features also relate to bias detection.
