Fairness-Constrained Reward Scorer

Software componentModel AdaptationModelsarc:FairnessConstrainedRewardScorer

A reward scorer that rewards decisions satisfying a selected group-fairness metric, such as equalized odds, alongside business-relevant criteria like debt-to-income ratio and credit history.

Responsibility. Rewards decisions for meeting a chosen fairness metric.

Also known as: Fairness-focused reward model

Variant of Reward Scorer abstract

When to choose. Choose for high-stakes decisions such as lending where the policy must satisfy a chosen fairness metric (e.g., equalized odds) while keeping legitimate business criteria.

Direct neighbourhood (hover for relationship types)

Relationships

reads dependency

Labeled Decision Case Dataset Ch9.5

alternative to variability

Constitutional Reward Scorer Ch9.5

Design guidance

MUST select the fairness metric by explicit value judgment; demographic parity and equalized odds mathematically conflict.
SHOULD NOT be relied on alone: historical data bias and proxy features require separate auditing and monitoring.

Classification

Patterns: Equalized odds optimization
Quality attributes: Fairness (NIST AI RMF: fair, harmful bias managed)Transparency and accountability (NIST AI RMF: accountable and transparent)
Risks mitigated: Disparate treatment across demographic groups
Frameworks & regulations: Equal Credit Opportunity ActFair Housing Act

Sources

Ch9.5: T. Nguyen, "Constitutional AI," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 9.5. ISBN: 9798244538229.