ProbLog4Fairness: A Neurosymbolic Approach to Modeling and Mitigating Bias

2025-11-12

AI Summary

Existing fairness definitions are often mutually incompatible and lack a principled, flexible, and interpretable framework capable of seamlessly integrating domain knowledge—such as assumptions about systemic biases—into model training. Method: We propose a neuro-symbolic approach that formalizes diverse bias hypotheses as interpretable, composable bias expression templates within a probabilistic logic program built on ProbLog, and tightly couples this symbolic representation with neural network training. Contribution/Results: Our method transcends conventional reliance on fixed bias types or singular fairness criteria, enabling synergistic causal reasoning and deep learning. It supports adaptive bias identification and mitigation on both tabular and image data. Experiments demonstrate statistically significant improvements over state-of-the-art baselines on synthetic and real-world datasets, confirming strong generalization and cross-domain applicability.

Technology Category

• Machine Learning: Neuro-Symbolic Learning
• Computer Vision: Visual Reasoning & Symbolic Representations
• 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
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Operationalizing definitions of fairness is difficult in practice, as multiple definitions can be incompatible while each being arguably desirable. Instead, it may be easier to directly describe algorithmic bias through ad-hoc assumptions specific to a particular real-world task, e.g., based on background information on systemic biases in its context. Such assumptions can, in turn, be used to mitigate this bias during training. Yet, a framework for incorporating such assumptions that is simultaneously principled, flexible, and interpretable is currently lacking. Our approach is to formalize bias assumptions as programs in ProbLog, a probabilistic logic programming language that allows for the description of probabilistic causal relationships through logic. Neurosymbolic extensions of ProbLog then allow for easy integration of these assumptions in a neural network's training process. We propose a set of templates to express different types of bias and show the versatility of our approach on synthetic tabular datasets with known biases. Using estimates of the bias distortions present, we also succeed in mitigating algorithmic bias in real-world tabular and image data. We conclude that ProbLog4Fairness outperforms baselines due to its ability to flexibly model the relevant bias assumptions, where other methods typically uphold a fixed bias type or notion of fairness.

Problem

• Modeling algorithmic bias through probabilistic logic programming
• Mitigating bias in neural networks using neurosymbolic integration
• Providing flexible framework for diverse fairness definitions

Innovation

• Uses ProbLog to formalize bias assumptions
• Integrates bias models into neural network training
• Provides templates for diverse bias types