用流模型生成稀疏、可解释的反事实样本，支持异构数据和约束条件。
CounterFlowNet: From Minimal Changes to Meaningful Counterfactual Explanations
- 基于条件生成流网络，逐步修改特征生成反事实解释。
- 在8个数据集上实现高有效性、稀疏性与多样性平衡，满足用户约束。
- 无需重训练即可处理不可变或单调性等实际限制，适合决策解释场景。
反事实解释（CFs）通过识别能改变模型输出的最小输入特征变更，提供人类可理解的决策洞察。现有方法难以生成多个高质量解释，且在仅修改少量特征、支持异构特征的表格数据、符合用户约束方面存在不足。本文提出CounterFlowNet，将CF生成建模为使用条件生成流网络（GFlowNet）的序列特征修改过程。该方法通过用户指定的奖励函数采样CF，以比例分配概率，编码有效性、稀疏性、接近性和合理性等关键需求，从而生成高质量解释。序列化设计实现高度稀疏的特征修改，统一的动作空间无缝支持连续与分类特征。此外，可通过动作掩码在推理时强制实施不可变性或单调性等可行性约束，无需重新训练。在八个数据集上，两种评估协议下均表现出更优的有效性、稀疏性、合理性与多样性权衡，完全满足给定约束。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。