DiCoFlex: Model-agnostic diverse counterfactuals with flexible control
This addresses the need for practical and scalable counterfactual explanations in sensitive decision-making domains, offering an incremental improvement over existing methods.
The paper tackled the problem of generating diverse counterfactual explanations for machine learning models, which often require model access and lack flexibility, by proposing DiCoFlex, a model-agnostic framework that produces multiple counterfactuals in a single forward pass with real-time user customization, achieving superior performance in validity, diversity, proximity, and constraint adherence on benchmark datasets.
Counterfactual explanations play a pivotal role in explainable artificial intelligence (XAI) by offering intuitive, human-understandable alternatives that elucidate machine learning model decisions. Despite their significance, existing methods for generating counterfactuals often require constant access to the predictive model, involve computationally intensive optimization for each instance and lack the flexibility to adapt to new user-defined constraints without retraining. In this paper, we propose DiCoFlex, a novel model-agnostic, conditional generative framework that produces multiple diverse counterfactuals in a single forward pass. Leveraging conditional normalizing flows trained solely on labeled data, DiCoFlex addresses key limitations by enabling real-time user-driven customization of constraints such as sparsity and actionability at inference time. Extensive experiments on standard benchmark datasets show that DiCoFlex outperforms existing methods in terms of validity, diversity, proximity, and constraint adherence, making it a practical and scalable solution for counterfactual generation in sensitive decision-making domains.