Unverified50% confidenceTradeoffExact time
Demographic parity, equal opportunity, and predictive parity often cannot be mathematically satisfied simultaneously (unless the true default rates are the same across groups and the model is perfect); forcing demographic parity distorts predicted probabilities and harms calibration, while pursuing predictive parity may accept group differences in approval rates.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedFairness is measured by Equalized Odds, a metric that requires the model to maintain consistent true positive rates and false positive rates across different demographic groups.76% similarUnverifiedWhen large numbers of users call the same underlying model with similar prompts, convergence of outputs is almost a mathematical inevitability due to the consistent underlying weights and probability distributions.70% similarUnverifiedFitting the propensity score in-sample induces dependence among individual contributions, so variance estimators derived under the i.i.d. assumption underestimate the true uncertainty.69% similarUnverifiedThe paper proves a matching minimax lower bound, showing that the Ω(δ²T) bias term is unavoidable — a bias cost must be paid whenever the guidance distribution differs from the model distribution.69% similarUnverifiedIn class-imbalance scenarios, a model can achieve seemingly impressive accuracy by always predicting the majority class, a phenomenon known as the Accuracy Paradox.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1074563API
curl https://kongchang.com/api/v1/knowledge/claims/1074563MCP
get_claim(id=1074563)