conftrace_

2024 AAAI AAAI 2024

Fair and Optimal Prediction via Post-Processing

Abstract

Abstract In this talk I will discuss our recent work on characterizing the inherent tradeoff between fairness and accuracy in both classification and regression problems. I will also present a post-processing algorithm that derives optimal fair predictors from Bayes score functions.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Mathematics & Optimization

Authors

Han Zhao

Topics

Artificial Intelligence > Core AI > Responsible AI Machine Learning > Core Methods > Classification Machine Learning > Core Methods > Regression Machine Learning > Application Areas > Fairness Mathematics & Optimization > Optimization > Continuous Optimization Artificial Intelligence > Core AI > Fairness

Keywords

fair prediction bayes score post-processing algorithm classification fairness fair predictor bayes score function fairness-accuracy trade-off optimal predictor

More from AAAI 2024

All AAAI 2024 papers →

Towards Reliable Learning in the Wild: Generalization and Adaptation

Towards Human-like Learning from Relational Structured Data

Fairness with Censorship: Bridging the Gap between Fairness Research and Real-World Deployment

Towards Reproducible, Automated, and Scalable Anomaly Detection

Combating Insider Threat in the Open-World Environments: Identification, Monitoring, and Data Augmentation

Select and Augment: Enhanced Dense Retrieval Knowledge Graph Augmentation (Abstract Reprint)