mlsanityData to 14 Oct 2026

Generalizability vs. Counterfactual Explainability Trade-Off

arXiv 2505.23225 ↗First posted 29 May 2025cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

In this work, we investigate the relationship between model generalization and counterfactual explainability in supervised learning. We introduce the notion of varepsilon-valid counterfactual probability (varepsilon-VCP) -- the probability of finding perturbations of a data point within its varepsilon-neighborhood that result in a label change. We provide a theoretical analysis of varepsilon-VCP in relation to the geometry of the model's decision boundary, showing that varepsilon-VCP tends to increase with model overfitting. Our findings establish a rigorous connection between poor generalization and the ease of counterfactual generation, revealing an inherent trade-off between generalization and counterfactual explainability. Empirical results validate our theory, suggesting varepsilon-VCP as a practical proxy for quantitatively characterizing overfitting.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 22 of its 36 references that are in the map. First posted on arXiv 29 May 2025.

Builds on

22 of 36 references.

References · 22

  1. 2015Domain-Adversarial Training of Neural Networkscited by 3,044
  2. 2018Learning deep representations by mutual information estimation and maximizationcited by 1,273
  3. 2017Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPRcited by 1,089
  4. 2017Exploring Generalization in Deep Learningcited by 600
  5. 2019Model-Agnostic Counterfactual Explanations for Consequential Decisionscited by 204
  6. 2018Lipschitz-Margin Training: Scalable Certification of Perturbation Invariance for Deep Neural Networkscited by 195
  7. 2017Formal Guarantees on the Robustness of a Classifier against Adversarial Manipulationcited by 185
  8. 2018Disentangling Adversarial Robustness and Generalizationcited by 172
  9. 2018Local Rule-Based Explanations of Black Box Decision Systemscited by 166
  10. 2019Learning Model-Agnostic Counterfactual Explanations for Tabular Datacited by 146
  11. 2021CF-GNNExplainer: Counterfactual Explanations for Graph Neural Networkscited by 119
  12. 2013Early stopping and non-parametric regression: An optimal data-dependent stopping rulecited by 89
10 more
  1. 2017Interpretable Predictions of Tree-based Ensembles via Actionable Feature Tweakingcited by 76
  2. 2020The Intriguing Relation Between Counterfactual Explanations and Adversarial Examplescited by 44
  3. 2019FOCUS: Flexible Optimizable Counterfactual Explanations for Tree Ensemblescited by 43
  4. 2021The Dimpled Manifold Model of Adversarial Examples in Machine Learningcited by 41
  5. 2018Unrestricted Adversarial Examplescited by 40
  6. 2020CounteRGAN: Generating Realistic Counterfactuals with Residual Generative Adversarial Netscited by 29
  7. 2021ReLAX: Reinforcement Learning Agent eXplainer for Arbitrary Predictive Modelscited by 25
  8. 2013An empirical analysis of dropout in piecewise linear networkscited by 19
  9. 2024Eliminating Catastrophic Overfitting Via Abnormal Adversarial Examples Regularizationcited by 12
  10. 2019Understanding Adversarial Robustness: The Trade-off between Minimum and Average Margincited by 5

Built on by

Not cited yet.