Yang Rui, Basin Michael V, Yao Guangzhe, Zeng Hongzheng
Robotics Institute, Ningbo University of Technology, Ningbo 315211, China.
School of Physical and Mathematical Sciences, Autonomous University of Nuevo León, San Nicolás de Los Garza 66455, Nuevo León, Mexico.
Sensors (Basel). 2025 Aug 12;25(16):4982. doi: 10.3390/s25164982.
Inspired by the interpretability of Kolmogorov-Arnold Networks (KANs), a novel Pixel-level Feature Selection (PFS) model based on KANs (PFSKANs) is proposed as a fundamentally distinct alternative from trainable Convolutional Neural Networks (CNNs) and transformers in the computer vision tasks. We modify the simplification techniques of KANs to detect key pixels with high contribution scores directly at the input image. Specifically, a trainable selection procedure is intuitively visualized and performed only once, since the obtained interpretable pixels can subsequently be identified and dimensionally standardized using the proposed mathematical approach. Experiments on the image classification tasks using the MNIST, Fashion-MNIST, CIFAR-10, and CIFAR-100 datasets demonstrate that PFSKANs achieve comparable performance to CNNs in terms of accuracy, parameter efficiency, and training time.
受柯尔莫哥洛夫 - 阿诺德网络（KANs）可解释性的启发，提出了一种基于KANs的新型像素级特征选择（PFS）模型（PFSKANs），作为计算机视觉任务中与可训练卷积神经网络（CNNs）和变压器截然不同的替代方案。我们修改了KANs的简化技术，以直接在输入图像上检测具有高贡献分数的关键像素。具体来说，一个可训练的选择过程直观地可视化并且只执行一次，因为随后可以使用所提出的数学方法识别获得的可解释像素并进行维度标准化。使用MNIST、Fashion - MNIST、CIFAR - 10和CIFAR - 100数据集进行的图像分类任务实验表明，PFSKANs在准确性、参数效率和训练时间方面与CNNs具有相当的性能。
