Bias Mitigation
Risk and SafetyEssentialAppliedNeeds source verification
Definition
Specific steps and techniques taken to identify and reduce bias in models and datasets.
Context
HR
Example
We will use a bias mitigation strategy to address fairness concerns in the model.
Governance implication
Identify who is responsible for applying and reviewing this concept in a healthcare AI workflow.
Question to ask
How does this concept affect the safety, reliability, or governance of the AI tool we use?
Relevant to
AI Attorney, AI Project Manager, HR, Healthcare Leader, Nurse, Pharmacist, Physician (Suggested — not reviewed)
Source: Excel column 2; ubitiquos_lang.pdf
Verification: Not independently verified
Version 0.1-draft · AI-GLOSS-0039