AI 中文总结
该研究针对结构化因果输入下神经网络预测解释的误导性问题，提出基于布尔结构因果模型与边界传播、分支定界技术的方法，可高效计算所有最小实际原因，性能优于多种基线方法。
AI 中文摘要
解释神经网络的预测结果是可信人工智能领域的核心挑战。现有解释方法，比如基于特征归因或最小充分集的方法，通常将输入特征视为独立的，而当输入存在结构化依赖关系时，这类方法会产生误导性的解释。为解决该问题，我们将解释形式化为Halpern-Pearl（HP）实际原因，使用布尔结构因果模型（SCM）对输入依赖关系进行建模。我们通过应用边界传播和分支定界技术来计算HP原因，同时提供完备性和极小性的形式保证。实验结果表明，我们的方法在可扩展性上显著优于暴力搜索和整数线性规划（ILP）基线，且随着图规模增长，性能优于启发式搜索；在每个实例180秒的时间预算内，我们在最多28个节点的SCM上，对包含多达2.3×10^13个（原因，偶然因素）候选对的实例，计算了所有最小实际原因。在案例研究中，我们进一步发现，忽略输入依赖关系会导致报告的原因数量增加，在我们的SCM下，其中14.9%是虚假原因。
英文摘要
Explaining the predictions of neural networks is a central challenge in trustworthy AI. Existing explanation methods, such as those based on feature attribution or minimal sufficient sets, typically treat input features as independent, which can yield misleading explanations when inputs exhibit structured dependencies. We address this by formalizing explanations as Halpern-Pearl (HP) actual causes, modeling input dependencies using Boolean Structural Causal Models (SCMs). We compute HP causes by applying bound propagation and branch-and-bound techniques, while providing formal guarantees of completeness and minimality. Our experiments show that we substantially outperform brute-force and ILP baselines in scalability, and outperform heuristic search as graph size grows, computing all minimal actual causes on instances with search spaces of up to $2.3\times10^{13}$ candidate (cause, contingency) pairs, on SCMs with up to 28 nodes, within a 180s per-instance budget. In a case study, we further show that ignoring input dependencies inflates the number of reported causes, 14.9% of which are spurious under our SCM.
