提出可解释的神经网络，精准捕捉表格数据中特征间配对交互。
Tree-like Pairwise Interaction Networks
- 用类决策树结构显式建模特征配对交互，提升可解释性。
- 在法国车险数据上，预测精度超越传统与现代神经网络模型。
- 适合需要透明决策过程的保险、金融等高风险领域使用。
表格数据中的特征交互建模仍是预测建模的关键挑战，例如在保险定价中的应用。本文提出树状配对交互网络（Tree-like Pairwise Interaction Network, PIN），一种新型神经网络架构，通过共享前馈网络结构显式捕捉配对特征交互，其结构模仿决策树。该模型天然具备可解释性，可直接观察交互效应。此外，由于仅涉及配对交互，能高效计算SHapley加性解释（SHAP）。我们揭示了PIN与已有模型（如GA2Ms、梯度提升机、图神经网络）的关联。在流行的法国车险数据集上的实验表明，PIN在预测准确率上优于传统及现代神经网络基准，同时揭示特征间的交互方式及其对预测的贡献。
原文摘要 · Abstract (English)
Modeling feature interactions in tabular data remains a key challenge in predictive modeling, for example, as used for insurance pricing. This paper proposes the Tree-like Pairwise Interaction Network (PIN), a novel neural network architecture that explicitly captures pairwise feature interactions through a shared feed-forward neural network architecture that mimics the structure of decision trees. PIN enables intrinsic interpretability by design, allowing for direct inspection of interaction effects. Moreover, it allows for efficient SHapley's Additive exPlanation (SHAP) computations because it only involves pairwise interactions. We highlight connections between PIN and established models such as GA2Ms, gradient boosting machines, and graph neural networks. Empirical results on the popular French motor insurance dataset show that PIN outperforms both traditional and modern neural networks benchmarks in predictive accuracy, while also providing insight into how features interact with each another and how they contribute to the predictions.
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。