Is Each Layer Non-trivial in CNN? (Student Abstract)

Authors

  • Wei Wang Shenzhen Institutes of Advanced Technology,Chinese Academy of Sciences
  • Yanjie Zhu Shenzhen Institutes of Advanced Technology,Chinese Academy of Sciences
  • Zhuoxu Cui Shenzhen Institutes of Advanced Technology,Chinese Academy of Sciences
  • Dong Liang shenzhen Institutes of Advanced Technology,Chinese Academy of Sciences

Keywords:

ResNet, Convolution Kernel, Deep Learning

Abstract

Convolutional neural network (CNN) models have achieved great success in many fields. With the advent of ResNet, networks used in practice are getting deeper and wider. However, is each layer non-trivial in networks? To answer this question, we trained a network on the training set, then we replace the network convolution kernels with zeros and test the result models on the test set. We compared experimental results with baseline and showed that we can reach similar or even the same performances. Although convolution kernels are the cores of networks, we demonstrate that some of them are trivial and regular in ResNet.

Downloads

Published

2021-05-18

How to Cite

Wang, W., Zhu, Y., Cui, Z., & Liang, D. (2021). Is Each Layer Non-trivial in CNN? (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence, 35(18), 15915-15916. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/17954

Issue

Section

AAAI Student Abstract and Poster Program