A new machine learning methodology, called successive subspace learning (SSL), is introduced in this work. SSL contains four key ingredients: (1) successive near-to-far neighborhood expansion; (2) unsupervised dimension reduction via subspace approximation; (3) supervised dimension reduction via label-assisted regression (LAG); and (4) feature concatenation and decision making. An image-based object classification method, called PixelHop, is proposed to illustrate the SSL design. It is shown by experimental results that the PixelHop method outperforms the classic CNN model of similar model complexity in three benchmarking datasets (MNIST, Fashion MNIST and CIFAR-10). Although SSL and deep learning (DL) have some high-level concept in common, they are fundamentally different in model formulation, the training process and training complexity. Extensive discussion on the comparison of SSL and DL is made to provide further insights into the potential of SSL.

在这项工作中引入了一种新的机器学习方法，称为连续子空间学习（SSL）。SSL包含四个关键要素：（1）连续的从近到远的邻域扩展；（2）通过子空间逼近的无监督降维；（3）通过标签辅助回归（LAG）进行有监督的尺寸缩减；（4）特征级联和决策。提出了一种基于图像的对象分类方法，称为PixelHop，以说明SSL设计。实验结果表明，在三个基准数据集（MNIST，Fashion MNIST和CIFAR-10）中，PixelHop方法优于具有类似模型复杂性的经典CNN模型。尽管SSL和深度学习（DL）有一些共同的高级概念，但是它们在模型制定，训练过程和训练复杂性方面根本不同。