A Critical Review of Predominant Bias in Neural Networks

2025-02-16

AI Summary

Persistent conceptual ambiguity in fair machine learning has conflated two distinct biases: inter-group performance disparity (e.g., violation of statistical parity) and implicit reliance on protected attributes (e.g., counterfactual unfairness), leading to flawed evaluation and methodological confusion. Method: We introduce the first unified mathematical framework formalizing both biases, construct an empirical confounding attribution system across synthetic, census, and image datasets, and establish a multidimensional comparative paradigm—spanning causal origins, mitigation methods, fairness metrics, and benchmarks. Using formal modeling, controlled experiments, and bias measurement matrices (SPD/EOD), we systematically analyze their divergence. Contribution/Results: Our analysis reveals fundamental distinctions between the two biases, identifies seven canonical confounding patterns, and exposes systematic failures of mainstream fairness metrics under cross-bias scenarios. The work delivers a reproducible conceptual clarification guide and an empirical benchmark for fair AI research.

Technology Category

• Machine Learning: Ethics, Bias, and Fairness
• Computer Vision: Bias, Fairness & Privacy
• Philosophy and Ethics of AI: Bias, Fairness & Equity

Application Category

• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Bias issues of neural networks garner significant attention along with its promising advancement. Among various bias issues, mitigating two predominant biases is crucial in advancing fair and trustworthy AI: (1) ensuring neural networks yields even performance across demographic groups, and (2) ensuring algorithmic decision-making does not rely on protected attributes. However, upon the investigation of pc papers in the relevant literature, we find that there exists a persistent, extensive but under-explored confusion regarding these two types of biases. Furthermore, the confusion has already significantly hampered the clarity of the community and subsequent development of debiasing methodologies. Thus, in this work, we aim to restore clarity by providing two mathematical definitions for these two predominant biases and leveraging these definitions to unify a comprehensive list of papers. Next, we highlight the common phenomena and the possible reasons for the existing confusion. To alleviate the confusion, we provide extensive experiments on synthetic, census, and image datasets, to validate the distinct nature of these biases, distinguish their different real-world manifestations, and evaluate the effectiveness of a comprehensive list of bias assessment metrics in assessing the mitigation of these biases. Further, we compare these two types of biases from multiple dimensions including the underlying causes, debiasing methods, evaluation protocol, prevalent datasets, and future directions. Last, we provide several suggestions aiming to guide researchers engaged in bias-related work to avoid confusion and further enhance clarity in the community.

Problem

• Addressing neural network performance disparities
• Eliminating reliance on protected attributes
• Clarifying and mitigating two predominant biases

Innovation

• Mathematical definitions for biases
• Experiments on diverse datasets
• Comparison of debiasing methods