ElliCE: Efficient and Provably Robust Algorithmic Recourse via the Rashomon Sets

2026-02-07

AI Summary

This work addresses the limited robustness of traditional counterfactual explanations, which are tailored to a single model and often fail to remain valid under model uncertainty or across near-optimal models. To overcome this, the authors propose a novel approach that constructs an ellipsoidal approximation of the Rashomon set and optimizes counterfactual generation over this set, marking the first integration of such an approximation into an algorithmic recourse framework. The method offers theoretical guarantees regarding uniqueness, stability, and alignment of feature directions, supports user-defined constraints, and enables efficient computation. Empirical results demonstrate that the generated counterfactuals remain effective across multiple near-optimal models, achieve significantly faster computation than existing baselines, and exhibit enhanced flexibility and robustness.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Reasoning under Uncertainty: Causality
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Machine learning models now influence decisions that directly affect people's lives, making it important to understand not only their predictions, but also how individuals could act to obtain better results. Algorithmic recourse provides actionable input modifications to achieve more favorable outcomes, typically relying on counterfactual explanations to suggest such changes. However, when the Rashomon set - the set of near-optimal models - is large, standard counterfactual explanations can become unreliable, as a recourse action valid for one model may fail under another. We introduce ElliCE, a novel framework for robust algorithmic recourse that optimizes counterfactuals over an ellipsoidal approximation of the Rashomon set. The resulting explanations are provably valid over this ellipsoid, with theoretical guarantees on uniqueness, stability, and alignment with key feature directions. Empirically, ElliCE generates counterfactuals that are not only more robust but also more flexible, adapting to user-specified feature constraints while being substantially faster than existing baselines. This provides a principled and practical solution for reliable recourse under model uncertainty, ensuring stable recommendations for users even as models evolve.

Problem

• algorithmic recourse
• counterfactual explanations
• Rashomon set
• model uncertainty
• robustness

Innovation

• algorithmic recourse
• Rashomon set
• counterfactual explanations
• ellipsoidal approximation
• model uncertainty