Post-processing for Individual Fairness

2021/10/26 by Felix Petersen, Debarghya Mukherjee, Petersen, Felix +5 · 6 citations
Social Sciences · Computer Science · #Ethics and Social Impacts of AI #Explainable Artificial Intelligence (XAI) #Privacy-Preserving Technologies in Data

paper · pdf · doi:10.48550/arxiv.2110.13796

Abstract

Post-processing in algorithmic fairness is a versatile approach for correcting bias in ML systems that are already used in production. The main appeal of post-processing is that it avoids expensive retraining. In this work, we propose general post-processing algorithms for individual fairness (IF). We consider a setting where the learner only has access to the predictions of the original model and a similarity graph between individuals, guiding the desired fairness constraints. We cast the IF post-processing problem as a graph smoothing problem corresponding to graph Laplacian regularization that preserves the desired "treat similar individuals similarly" interpretation. Our theoretical results demonstrate the connection of the new objective function to a local relaxation of the original individual fairness. Empirically, our post-processing algorithms correct individual biases in large-scale NLP models such as BERT, while preserving accuracy.

Cited by

• FairExpand: Individual Fairness on Graphs with Partial Similarity Information
• FairAgent: Democratizing Fairness-Aware Machine Learning with LLM-Powered Agents
• A Multi-Objective Evaluation Framework for Analyzing Utility-Fairness Trade-Offs in Machine Learning Systems
• 'Rich Dad, Poor Lad': How do Large Language Models Contextualize Socioeconomic Factors in College Admission ?
• Machine Learning with Multitype Protected Attributes: Intersectional Fairness through Regularisation
• SaGIF: Improving Individual Fairness in Graph Neural Networks via Similarity Encoding

Related

• Explainable post-training bias mitigation with distribution-based fairness metrics
• Impact of Data Processing on Fairness in Supervised Learning
• Individual Fairness under Uncertainty
• iFlipper: Label Flipping for Individual Fairness
• Individual Fairness Through Reweighting and Tuning