The Pareto frontier of fair algorithmic decisions consists of deterministic group-specific threshold rules on predicted success probabilities, which can include upper bounds for some fairness metrics and holds independently of model training approach.
Equality of Opportunity in Supervised Learning
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
abstract
We propose a criterion for discrimination against a specified sensitive attribute in supervised learning, where the goal is to predict some target based on available features. Assuming data about the predictor, target, and membership in the protected group are available, we show how to optimally adjust any learned predictor so as to remove discrimination according to our definition. Our framework also improves incentives by shifting the cost of poor classification from disadvantaged groups to the decision maker, who can respond by improving the classification accuracy. In line with other studies, our notion is oblivious: it depends only on the joint statistics of the predictor, the target and the protected attribute, but not on interpretation of individualfeatures. We study the inherent limits of defining and identifying biases based on such oblivious measures, outlining what can and cannot be inferred from different oblivious tests. We illustrate our notion using a case study of FICO credit scores.
citation-role summary
extension 1
citation-polarity summary
roles
extension 1polarities
extend 1representative citing papers
Error-rate balance and predictive parity become compatible under endogenous behavior by adjusting stakes differently across groups, introducing a new form of unequal treatment in consequences.
The authors provide a detailed taxonomy of 21 risks associated with language models, covering discrimination, information leaks, misinformation, malicious applications, interaction harms, and societal impacts like job loss and environmental costs.
Develops a bandit algorithm with graph feedback that learns weights for multiple fairness constraints adaptively over sequential interactions.
Vision-language models for wellbeing assessment exhibit dataset-dependent performance and demographic biases, with explainability interventions providing inconsistent fairness gains at potential accuracy costs.
A synthetic review across multiple fields concludes that AI decision aids have modest or nonexistent effects on judicial outcomes while identifying gaps in understanding human-AI interactions.
citing papers explorer
-
Fairness vs Performance: Characterizing the Pareto Frontier of Algorithmic Decision Systems
The Pareto frontier of fair algorithmic decisions consists of deterministic group-specific threshold rules on predicted success probabilities, which can include upper bounds for some fairness metrics and holds independently of model training approach.
-
Revisiting Fairness Impossibility with Endogenous Behavior
Error-rate balance and predictive parity become compatible under endogenous behavior by adjusting stakes differently across groups, introducing a new form of unequal treatment in consequences.
-
Ethical and social risks of harm from Language Models
The authors provide a detailed taxonomy of 21 risks associated with language models, covering discrimination, information leaks, misinformation, malicious applications, interaction harms, and societal impacts like job loss and environmental costs.
-
Online Learning with Multiple Fairness Regularizers via Graph-Structured Feedback
Develops a bandit algorithm with graph feedback that learns weights for multiple fairness constraints adaptively over sequential interactions.
-
FAIR_XAI: Improving Multimodal Foundation Model Fairness via Explainability for Wellbeing Assessment
Vision-language models for wellbeing assessment exhibit dataset-dependent performance and demographic biases, with explainability interventions providing inconsistent fairness gains at potential accuracy costs.
-
Man and machine: artificial intelligence and judicial decision making
A synthetic review across multiple fields concludes that AI decision aids have modest or nonexistent effects on judicial outcomes while identifying gaps in understanding human-AI interactions.