CounterFlowNet: From Minimal Changes to Meaningful Counterfactual Explanations

2026-02-19

AI Summary

This work proposes CounterFlowNet, a novel approach to generating high-quality counterfactual explanations that simultaneously satisfy sparsity, compatibility with heterogeneous features, and user-specified constraints—challenges that existing methods struggle to address collectively. CounterFlowNet introduces conditional generative flow networks (GFlowNets) to counterfactual generation, modeling explanation construction as a sequential feature-editing process. It employs an action masking mechanism to uniformly handle both continuous and discrete features and integrates a user-defined reward function encompassing validity, sparsity, proximity, and plausibility. Crucially, operational constraints can be flexibly enforced during inference without retraining. Extensive experiments across eight datasets demonstrate that CounterFlowNet achieves a superior trade-off among validity, sparsity, plausibility, and diversity while strictly adhering to user-imposed constraints.

Technology Category

• Machine Learning: Feature Construction/Reformulation
• Humans and AI: Explainable AI (XAI) for Human Understanding
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Counterfactual explanations (CFs) provide human-interpretable insights into model's predictions by identifying minimal changes to input features that would alter the model's output. However, existing methods struggle to generate multiple high-quality explanations that (1) affect only a small portion of the features, (2) can be applied to tabular data with heterogeneous features, and (3) are consistent with the user-defined constraints. We propose CounterFlowNet, a generative approach that formulates CF generation as sequential feature modification using conditional Generative Flow Networks (GFlowNet). CounterFlowNet is trained to sample CFs proportionally to a user-specified reward function that can encode key CF desiderata: validity, sparsity, proximity and plausibility, encouraging high-quality explanations. The sequential formulation yields highly sparse edits, while a unified action space seamlessly supports continuous and categorical features. Moreover, actionability constraints, such as immutability and monotonicity of features, can be enforced at inference time via action masking, without retraining. Experiments on eight datasets under two evaluation protocols demonstrate that CounterFlowNet achieves superior trade-offs between validity, sparsity, plausibility, and diversity with full satisfaction of the given constraints.

Problem

• counterfactual explanations
• sparsity
• tabular data
• actionability constraints
• heterogeneous features

Innovation

• Counterfactual Explanations
• Generative Flow Networks
• Sequential Feature Modification
• Action Masking
• Tabular Data