UGCE: User-Guided Incremental Counterfactual Exploration

TL;DR

This work tackles the problem of generating counterfactual explanations under dynamically evolving user constraints. It introduces UGCE, a genetic-algorithm–based framework that incrementally updates counterfactuals by reusing and repairing the previously evolved population instead of restarting from scratch. Through extensive experiments on five benchmark datasets, UGCE demonstrates substantial reductions in computation time while maintaining high-quality counterfactuals and stability across diverse constraint updates. The results highlight UGCE's practicality for real-time, interactive explainability in high-stakes domains and point to promising directions for dynamic weighting, user feedback integration, and domain-specific feasibility priors.

Abstract

Counterfactual explanations (CFEs) are a popular approach for interpreting machine learning predictions by identifying minimal feature changes that alter model outputs. However, in real-world settings, users often refine feasibility constraints over time, requiring counterfactual generation to adapt dynamically. Existing methods fail to support such iterative updates, instead recomputing explanations from scratch with each change, an inefficient and rigid approach. We propose User-Guided Incremental Counterfactual Exploration (UGCE), a genetic algorithm-based framework that incrementally updates counterfactuals in response to evolving user constraints. Experimental results across five benchmark datasets demonstrate that UGCE significantly improves computational efficiency while maintaining high-quality solutions compared to a static, non-incremental approach. Our evaluation further shows that UGCE supports stable performance under varying constraint sequences, benefits from an efficient warm-start strategy, and reveals how different constraint types may affect search behavior.

Figures (1)

• Figure 1: UGCE Pipeline