BM-CL: Bias Mitigation through the lens of Continual Learning

2025-09-01

AI Summary

Machine learning models often amplify biases against disadvantaged groups, while conventional debiasing methods suffer from the “performance drop effect”—sacrificing majority-group accuracy to improve fairness. This paper reformulates bias mitigation as a domain-incremental continual learning task and proposes a progressive debiasing framework with dynamically adjusted fairness objectives. By integrating continual learning techniques—including Elastic Weight Consolidation (EWC) and Learning without Forgetting (LwF)—the framework jointly optimizes fairness improvement and preservation of pre-existing knowledge. Experiments on synthetic and real-world image datasets demonstrate that our approach significantly reduces inter-group prediction disparities (e.g., up to 42% reduction in equal opportunity difference) while constraining accuracy loss on majority groups to within 0.3%. This effectively alleviates the fairness–accuracy trade-off—a longstanding bottleneck in algorithmic fairness.

Technology Category

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

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Biases in machine learning pose significant challenges, particularly when models amplify disparities that affect disadvantaged groups. Traditional bias mitigation techniques often lead to a {itshape leveling-down effect}, whereby improving outcomes of disadvantaged groups comes at the expense of reduced performance for advantaged groups. This study introduces Bias Mitigation through Continual Learning (BM-CL), a novel framework that leverages the principles of continual learning to address this trade-off. We postulate that mitigating bias is conceptually similar to domain-incremental continual learning, where the model must adjust to changing fairness conditions, improving outcomes for disadvantaged groups without forgetting the knowledge that benefits advantaged groups. Drawing inspiration from techniques such as Learning without Forgetting and Elastic Weight Consolidation, we reinterpret bias mitigation as a continual learning problem. This perspective allows models to incrementally balance fairness objectives, enhancing outcomes for disadvantaged groups while preserving performance for advantaged groups. Experiments on synthetic and real-world image datasets, characterized by diverse sources of bias, demonstrate that the proposed framework mitigates biases while minimizing the loss of original knowledge. Our approach bridges the fields of fairness and continual learning, offering a promising pathway for developing machine learning systems that are both equitable and effective.

Problem

• Mitigating biases without disadvantaging advantaged groups
• Addressing fairness trade-offs through continual learning
• Balancing fairness objectives while preserving model performance

Innovation

• Leverages continual learning for bias mitigation
• Applies domain-incremental techniques to fairness
• Preserves performance while enhancing disadvantaged outcomes