Spline-KAN比MLP更快收敛，适合非独立同分布数据的联邦学习。
Evaluating Federated Kolmogorov-Arnold Networks on Non-IID Data
- 用样条函数构建KAN，替代传统MLP进行联邦学习
- Spline-KAN在50轮内达到MLP 100轮的准确率
- 适合追求训练效率的联邦学习研究者
Federated Kolmogorov-Arnold Networks (F-KANs) 已被提出，但其评估仍处于初期阶段。我们比较了使用样条函数和径向基函数作为激活函数的KAN与参数量相近的多层感知机（MLPs），在包含100个客户端的非独立同分布（non-IID）MNIST分类任务中进行100轮联邦学习。每个模型经过15次实验后发现，最优准确率可由样条KAN（Spline-KAN）在50轮内实现，而相同准确率需MLP耗时100轮，仅带来适度的计算时间增加。
原文摘要 · Abstract (English)
Federated Kolmogorov-Arnold Networks (F-KANs) have already been proposed, but their assessment is at an initial stage. We present a comparison between KANs (using B-splines and Radial Basis Functions as activation functions) and Multi- Layer Perceptrons (MLPs) with a similar number of parameters for 100 rounds of federated learning in the MNIST classification task using non-IID partitions with 100 clients. After 15 trials for each model, we show that the best accuracies achieved by MLPs can be achieved by Spline-KANs in half of the time (in rounds), with just a moderate increase in computing time.
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。