Kolmogorov-Arnold 网络在表格数据上的基准测试研究
A Benchmarking Study of Kolmogorov-Arnold Networks on Tabular Data
AI总结:
本文对 Kolmogorov-Arnold 网络(KANs)与多层感知器(MLPs)在真实表格数据集上进行基准测试,发现 KANs 在准确率和 F1 分数上优于或相当,尤其在数据量大的场景下,但计算成本更高。
摘要:
Kolmogorov-Arnold 网络(KANs)最近被引入机器学习领域,迅速吸引了整个社区的关注。然而,KANs 主要用于逼近复杂函数或处理合成数据,目前缺乏在真实世界表格数据集上的测试。本文提出了一项基准测试研究,比较 KANs 和多层感知器(MLPs)在表格数据集上的表现。该研究评估了任务性能和训练时间。从各个数据集上获得的结果来看,KANs 表现出优于或相当的性能,在准确率和 F1 分数上尤其擅长处理具有大量实例的数据集,表明其能稳健地处理复杂数据。我们还强调,与同等规模的 MLPs 相比,KANs 的性能提升伴随着更高的计算成本。
英文摘要:
Kolmogorov-Arnold Networks (KANs) have very recently been introduced into the world of machine learning, quickly capturing the attention of the entire community. However, KANs have mostly been tested for approximating complex functions or processing synthetic data, while a test on real-world tabular datasets is currently lacking. In this paper, we present a benchmarking study comparing KANs and Multi-Layer Perceptrons (MLPs) on tabular datasets. The study evaluates task performance and training times. From the results obtained on the various datasets, KANs demonstrate superior or comparable accuracy and F1 scores, excelling particularly in datasets with numerous instances, suggesting robust handling of complex data. We also highlight that this performance improvement of KANs comes with a higher computational cost when compared to MLPs of comparable sizes.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。