LTBs-KAN：线性时间B-样条Kolmogorov-Arnold网络
LTBs-KAN: Linear-Time B-splines Kolmogorov-Arnold Networks
AI总结：
本文提出LTBs-KAN，通过线性复杂度的B-样条函数计算，提升KAN的效率和参数减少，实验证明其在MNIST、Fashion-MNIST和CIFAR-10上表现优异。
摘要：
Kolmogorov-Arnold Networks (KANs) 是一种近期提出的神经网络架构，提供了一种替代多层感知机 (MLPs) 的选择，具有改进的可解释性和表达性。然而，由于B-样条函数计算的递归性质，KANs比MLPs慢很多，限制了其应用。本文通过提出一种新型基础-样条线性时间B-样条Kolmogorov-Arnold网络 (LTBs-KAN) 来解决这些问题，具有线性复杂度。与以往依赖Boor-Mansfield-Cox样条算法或其他计算密集型数学函数的方法不同，我们的方法显著减少了计算负担。此外，我们进一步通过前向传递中的乘积和矩阵分解来减少模型的参数，而不牺牲性能。在MNIST、Fashion-MNIST和CIFAR-10上的实验表明，当用作建筑模块时，LTBs-KAN在时间复杂度和参数减少方面表现良好，相较于其他KAN实现。
英文摘要：
Kolmogorov-Arnold Networks (KANs) are a recent neural network architecture offering an alternative to Multilayer Perceptrons (MLPs) with improved explainability and expressibility. However, KANs are significantly slower than MLPs due to the recursive nature of B-spline function computations, limiting their application. This work addresses these issues by proposing a novel base-spline Linear-Time B-splines Kolmogorov-Arnold Network (LTBs-KAN) with linear complexity. Unlike previous methods that rely on the Boor-Mansfield-Cox spline algorithm or other computationally intensive mathematical functions, our approach significantly reduces the computational burden. Additionally, we further reduce model's parameter through product-of-sums matrix factorization in the forward pass without sacrificing performance. Experiments on MNIST, Fashion-MNIST and CIFAR-10 demonstrate that LTBs-KAN achieves good time complexity and parameter reduction, when used as building architectural blocks, compared to other KAN implementations.
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。
