CG-FKAN:面向通信受限环境的压缩网格联邦 Kolmogorov-Arnold 网络
CG-FKAN: Compressed-Grid Federated Kolmogorov-Arnold Networks for Communication Constrained Environment
AI总结:
针对联邦学习中 KAN 网格扩展导致的通信开销问题,本文提出 CG-FKAN 方法,通过稀疏化并仅传输必要系数来压缩网格,在通信受限下实现最多 13.6% 的 RMSE 降低,并推导了近似误差理论上界。
摘要:
联邦学习(FL)广泛应用于隐私关键应用,但存在可解释性受限的问题,而 Kolmogorov-Arnold Networks(KAN)通过可学习的样条函数解决了该限制。然而,现有应用 KAN 的 FL 研究忽略了网格扩展引入的通信开销,而该扩展对建模复杂函数至关重要。本文提出 CG-FKAN,在通信预算下通过稀疏化并仅传输必要系数来压缩扩展网格。实验表明,在通信受限设置下,CG-FKAN 的 RMSE 比固定网格 KAN 最多低 13.6%。此外,我们推导了其近似误差的理论上界。
英文摘要:
Federated learning (FL), widely used in privacy-critical applications, suffers from limited interpretability, whereas Kolmogorov-Arnold Networks (KAN) address this limitation via learnable spline functions. However, existing FL studies applying KAN overlook the communication overhead introduced by grid extension, which is essential for modeling complex functions. In this letter, we propose CG-FKAN, which compresses extended grids by sparsifying and transmitting only essential coefficients under a communication budget. Experiments show that CG-FKAN achieves up to 13.6% lower RMSE than fixed-grid KAN in communication-constrained settings. In addition, we derive a theoretical upper bound on its approximation error.
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。