The Limits of Group Fairness and Predictive Multiplicity

The Limits of Group Fairness and Predictive Multiplicity

Explore the boundaries of group fairness and predictive multiplicity in machine learning through this insightful lecture by Flavio Calmon from Harvard University. Delve into information-theoretic methods for developing trustworthy machine learning systems, examining the challenges and limitations of achieving fairness across different groups. Gain a deeper understanding of how multiple predictive models can impact decision-making processes and the implications for creating more equitable AI systems.