用可验证方法精确计算神经网络的SHAP值，突破了传统方法的规模瓶颈。
Verified SHAP: Provable Bounds for Exact Shapley Values of Neural Networks
- 利用神经网络验证技术，构建SHAP值的上下界
- 在更大输入空间上实现比现有方法快数个数量级的精确计算
- 为评估近似方法提供理论基准，适合关注模型可解释性的研究者
Shapley加性解释（SHAP）被广泛认为对神经网络计算不可行，因其需遍历输入特征的指数级搜索空间。本文首次通过引入新算法，利用神经网络验证的最新进展，为神经网络的SHAP值计算任意紧密的精确下界和上界，最终恢复出确切的SHAP值。实验表明，该方法可扩展至比当前最优精确方法大数个数量级的搜索空间，是迈向精确SHAP计算的重要一步，并为在更大搜索空间上评估统计近似方法提供了严谨的基准。
原文摘要 · Abstract (English)
Shapley additive explanations (SHAP) are widely recognised as computationally intractable for neural networks, since they induce an exponential search space over the input features. In this work, we take a first step towards scaling exact SHAP computation to larger search spaces by introducing an algorithm that leverages recent advances in neural network verification to compute arbitrarily tight exact lower and upper bounds on SHAP values for neural networks, ultimately recovering the exact SHAP values. We demonstrate that our approach scales to orders of magnitude larger search spaces than state-of-the-art exact methods. This provides an important first step towards exact SHAP computation and establishes a principled cornerstone for evaluating statistical approximation methods on larger search spaces.
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。