PixelHop: A successive subspace learning (SSL) method for object recognition |
| |
| Affiliation: | 1. Department of Computer Science, Jinan University, Guangzhou, China;2. Department of Computer Science and Information Engineering, National Dong Hwa University, Taiwan;1. Department of Business Administration, Wonkwang University, 460 Iksandae-ro, Iksan, Jeonbuk, South Korea;2. Engineering Research Center on Cloud Computing & Internet of Things and E-commerce Intelligence of Fujian Universities Quanzhou Normal University, No. 398, Donghai Street, Fengze District, Quanzhou 362000, China;3. School of Economics and Management, Xinyu University, No. 2666, Yangguang Street, Xinyu 338004, China;1. Department of Electronic and Information Engineering, Beijing Jiaotong University, Beijing 100044, China;2. College of information, Liaoning University, Liaoning 110036, China |
| |
| Abstract: | 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. |
| |
| Keywords: | Machine learning Subspace learning Computer vision Pattern recognition |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
