统计优势是否值得付出成本？KANs和MLPs在结构化数据分类中的实证比较
Is the Statistical Advantage Worth the Cost? An Empirical Comparison of KANs and MLPs for Structured Data Classification
浏览论文内容
中文总结 AI 辅助
该研究对KANs和MLPs在结构化表格分类任务上进行实证比较，在标准化设置下训练并评估性能。结果显示KANs在二元和多类领域统计上更优，但效应量表明其有更高复杂度。结论是KANs适用于高精度应用，MLPs对资源受限环境更合适，未来应扩展分析。
中文摘要 AI 辅助
本研究对Kolmogorov-Arnold网络（KANs）和多层感知器（MLPs）在结构化表格分类任务上进行了实证基准比较。鉴于对KANs作为替代函数逼近架构的兴趣日增，我们评估了它们在涵盖二元、多类、多标签和有序问题的12个公开可用数据集上的开箱即用性能。两个模型在标准化预处理、架构和固定超参数设置下训练，用测试准确率、F1分数、配对假设检验和效应量分析评估性能。结果表明，KANs在二元和多类领域在统计上优于MLPs，在所有数据集上有显著总体优势。然而，观察到的中等效应量（d = -0.46）引发了重要的成本效益考量：KANs通过基于自适应样条的映射提供卓越泛化能力，但相对于MLP基线，其参数和计算复杂度大幅更高。这些发现表明，KANs是高精度应用的首选，而MLPs对资源受限环境仍是稳健且高效的选择。未来工作应将此分析扩展到其他数据模态以进一步完善这些架构选择标准。
英文摘要
This study presents an empirical benchmarking comparison between Kolmogorov-Arnold Networks (KANs) and Multi-Layer Perceptrons (MLPs) on structured tabular classification tasks. Motivated by the growing interest in KANs as an alternative function-approximating architecture, we evaluate their out-of-the-box performance on twelve publicly available datasets spanning binary, multiclass, multilabel, and ordinal problems. Both models were trained under standardized preprocessing, architecture, and fixed hyperparameter settings, with performance assessed using test accuracy and F1-Score, paired hypothesis testing, and effect size analysis. Results show that KANs statistically outperform MLPs in binary and multiclass domains and achieve a significant aggregate advantage across all datasets. However, the observed medium effect size (d = -0.46) raises an important cost-benefit consideration: while KANs offer superior generalization through adaptive spline-based mappings, this advantage comes with substantially higher parameter and computational complexity relative to the MLP baseline. These findings suggest KANs are the preferred choice for high-precision applications, while MLPs remain a robust and efficient option for resource-constrained environments. Future work should extend this analysis to additional data modalities to further refine these architectural selection criteria.
发表机构
- Institute of Computer Science, University of the Philippines Los Baños（菲律宾大学洛斯巴尼奥斯分校计算机科学研究所）
- Machine Learning and Artificial Intelligence Applications Lab, University of the Philippines Los Baños（菲律宾大学洛斯巴尼奥斯分校机器学习与人工智能应用实验室）
机构由 AI 辅助整理，请以论文原文为准。
