Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
8953584 | Neurocomputing | 2018 | 17 Pages |
Abstract
The principal component analysis network (PCANet), which is one of the recently proposed deep learning architectures, achieves the state-of-the-art classification accuracy in various databases. However, the visualization or explanation of the PCANet is lacked. In this paper, we try to explain why PCANet works well from energy perspective point of view based on a set of experiments. The paper shows that the error rate of PCANet is qualitatively correlated with the inverse of the logarithm of BlockEnergy, which is the energy after the block sliding process of PCANet, and also this relation is quantified by using curve fitting method. The proposed energy explanation approach can also be used as a testing method for checking if every step of the constructed networks is necessary.
Related Topics
Physical Sciences and Engineering
Computer Science
Artificial Intelligence
Authors
Wu Jiasong, Qiu Shijie, Kong Youyong, Jiang Longyu, Chen Yang, Yang Wankou, Lotfi Senhadji, Shu Huazhong,