# The Hidden Assumptions Behind Counterfactual Explanations and Principal Reasons

arXiv: https://arxiv.org/abs/1912.04930 · HTML: https://mlsanity.com/paper/1912.04930/

## Key facts

- Cited by 115 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Dec 2019.

## Built on by (most cited first, 30 of 115)

- [OpenXAI: Towards a Transparent Evaluation of Model Explanations](https://mlsanity.com/paper/2206.11104/) (2022)
- [CF-GNNExplainer: Counterfactual Explanations for Graph Neural Networks](https://mlsanity.com/paper/2102.03322/) (2021)
- [Counterfactual Explanations Can Be Manipulated](https://mlsanity.com/paper/2106.02666/) (2021)
- [Towards Robust and Reliable Algorithmic Recourse](https://mlsanity.com/paper/2102.13620/) (2021)
- [Diffusion Visual Counterfactual Explanations](https://mlsanity.com/paper/2210.11841/) (2022)
- [The Fallacy of AI Functionality](https://mlsanity.com/paper/2206.09511/) (2022)
- [If Only We Had Better Counterfactual Explanations: Five Key Deficits to Rectify in the Evaluation of Counterfactual XAI Techniques](https://mlsanity.com/paper/2103.01035/) (2021)
- [Leveraging Sparse Linear Layers for Debuggable Deep Networks](https://mlsanity.com/paper/2105.04857/) (2021)
- [Which Explanation Should I Choose? A Function Approximation Perspective to Characterizing Post Hoc Explanations](https://mlsanity.com/paper/2206.01254/) (2022)
- [NICE: An Algorithm for Nearest Instance Counterfactual Explanations](https://mlsanity.com/paper/2104.07411/) (2021)
- [Counterfactuals and Causability in Explainable Artificial Intelligence: Theory, Algorithms, and Applications](https://mlsanity.com/paper/2103.04244/) (2021)
- [Explaining Black-Box Algorithms Using Probabilistic Contrastive Counterfactuals](https://mlsanity.com/paper/2103.11972/) (2021)
- [Post-Hoc Explanations Fail to Achieve their Purpose in Adversarial Contexts](https://mlsanity.com/paper/2201.10295/) (2022)
- [Exploring Counterfactual Explanations Through the Lens of Adversarial Examples: A Theoretical and Empirical Analysis](https://mlsanity.com/paper/2106.09992/) (2021)
- [Rethinking Explainability as a Dialogue: A Practitioner's Perspective](https://mlsanity.com/paper/2202.01875/) (2022)
- [Robust Counterfactual Explanations for Tree-Based Ensembles](https://mlsanity.com/paper/2207.02739/) (2022)
- [Towards the Unification and Robustness of Perturbation and Gradient Based Explanations](https://mlsanity.com/paper/2102.10618/) (2021)
- [Local Explanations via Necessity and Sufficiency: Unifying Theory and Practice](https://mlsanity.com/paper/2103.14651/) (2021)
- [Robust Counterfactual Explanations for Neural Networks With Probabilistic Guarantees](https://mlsanity.com/paper/2305.11997/) (2023)
- [Consistent Counterfactuals for Deep Models](https://mlsanity.com/paper/2110.03109/) (2021)
- [Optimal Counterfactual Explanations in Tree Ensembles](https://mlsanity.com/paper/2106.06631/) (2021)
- [On the Robustness of Sparse Counterfactual Explanations to Adverse Perturbations](https://mlsanity.com/paper/2201.09051/) (2022)
- [DualCF: Efficient Model Extraction Attack from Counterfactual Explanations](https://mlsanity.com/paper/2205.06504/) (2022)
- [On the Privacy Risks of Algorithmic Recourse](https://mlsanity.com/paper/2211.05427/) (2022)
- [Sparse Visual Counterfactual Explanations in Image Space](https://mlsanity.com/paper/2205.07972/) (2022)
- [Selective Explanations: Leveraging Human Input to Align Explainable AI](https://mlsanity.com/paper/2301.09656/) (2023)
- [GLOBE-CE: A Translation-Based Approach for Global Counterfactual Explanations](https://mlsanity.com/paper/2305.17021/) (2023)
- [Formalising the Robustness of Counterfactual Explanations for Neural Networks](https://mlsanity.com/paper/2208.14878/) (2022)
- [Counterfactual Explanations for Arbitrary Regression Models](https://mlsanity.com/paper/2106.15212/) (2021)
- [Counterfactual Shapley Additive Explanations](https://mlsanity.com/paper/2110.14270/) (2021)

## Abstract (arXiv)

Counterfactual explanations are gaining prominence within technical, legal, and business circles as a way to explain the decisions of a machine learning model. These explanations share a trait with the long-established "principal reason" explanations required by U.S. credit laws: they both explain a decision by highlighting a set of features deemed most relevant--and withholding others. These "feature-highlighting explanations" have several desirable properties: They place no constraints on model complexity, do not require model disclosure, detail what needed to be different to achieve a different decision, and seem to automate compliance with the law. But they are far more complex and subjective than they appear. In this paper, we demonstrate that the utility of feature-highlighting explanations relies on a number of easily overlooked assumptions: that the recommended change in feature values clearly maps to real-world actions, that features can be made commensurate by looking only at the distribution of the training data, that features are only relevant to the decision at hand, and that the underlying model is stable over time, monotonic, and limited to binary outcomes. We then explore several consequences of acknowledging and attempting to address these assumptions, including a paradox in the way that feature-highlighting explanations aim to respect autonomy, the unchecked power that feature-highlighting explanations grant decision makers, and a tension between making these explanations useful and the need to keep the model hidden. While new research suggests several ways that feature-highlighting explanations can work around some of the problems that we identify, the disconnect between features in the model and actions in the real world--and the subjective choices necessary to compensate for this--must be understood before these techniques can be usefully implemented.