MC3G: Model Agnostic Causally Constrained Counterfactual Generation

2025-08-24

AI Summary

In high-stakes decision-making, there exists a fundamental tension among model transparency, algorithmic privacy preservation, and the operational feasibility of counterfactual explanations. Method: We propose a model-agnostic, causally constrained counterfactual generation framework. It first constructs an interpretable rule-based surrogate model to approximate the black-box predictor; then incorporates a causal graph to identify critical feature dependencies; and finally explicitly models and optimizes user effort cost during counterfactual search. Contribution/Results: Unlike existing approaches, our framework generates counterfactuals that are more realistic, fair, and low-effort—while preserving the intellectual property of proprietary models. It achieves significant improvements over baseline methods in interpretability, actionability, and computational efficiency. Extensive experiments on multiple real-world datasets validate its effectiveness.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Responsible Web: Algorithmic accountability and transparency on the web

Abstract

Machine learning models increasingly influence decisions in high-stakes settings such as finance, law and hiring, driving the need for transparent, interpretable outcomes. However, while explainable approaches can help understand the decisions being made, they may inadvertently reveal the underlying proprietary algorithm: an undesirable outcome for many practitioners. Consequently, it is crucial to balance meaningful transparency with a form of recourse that clarifies why a decision was made and offers actionable steps following which a favorable outcome can be obtained. Counterfactual explanations offer a powerful mechanism to address this need by showing how specific input changes lead to a more favorable prediction. We propose Model-Agnostic Causally Constrained Counterfactual Generation (MC3G), a novel framework that tackles limitations in the existing counterfactual methods. First, MC3G is model-agnostic: it approximates any black-box model using an explainable rule-based surrogate model. Second, this surrogate is used to generate counterfactuals that produce a favourable outcome for the original underlying black box model. Third, MC3G refines cost computation by excluding the ``effort" associated with feature changes that occur automatically due to causal dependencies. By focusing only on user-initiated changes, MC3G provides a more realistic and fair representation of the effort needed to achieve a favourable outcome. We show that MC3G delivers more interpretable and actionable counterfactual recommendations compared to existing techniques all while having a lower cost. Our findings highlight MC3G's potential to enhance transparency, accountability, and practical utility in decision-making processes that incorporate machine-learning approaches.

Problem

• Generates counterfactual explanations for black-box model decisions
• Balances transparency with algorithm protection in high-stakes domains
• Computes effort costs only for user-initiated feature changes

Innovation

• Model-agnostic rule-based surrogate approximation
• Causally constrained counterfactual generation
• User-initiated effort cost computation