Topological optimization of the DenseNet with pretrained-weights inheritance and genetic channel selection

作者：

Highlights：

• A GA training pipeline to select key input channels of DenseNet automatically.

• A ’Weight inheritance’ strategy to improve efficiency whilst maintaining accuracy.

• To reduce up to 30% parameters yet achieving a similar accuracy as the DenseNet.

• Validated robustness/scalability on several image recognition/classification datasets.

摘要

•A GA training pipeline to select key input channels of DenseNet automatically.•A ’Weight inheritance’ strategy to improve efficiency whilst maintaining accuracy.•To reduce up to 30% parameters yet achieving a similar accuracy as the DenseNet.•Validated robustness/scalability on several image recognition/classification datasets.

论文关键词：Deep convolutional neural networks,Genetic algorithms,Parameter reduction,Structure optimization,DenseNet

论文评审过程：Received 6 March 2020, Revised 9 July 2020, Accepted 19 August 2020, Available online 22 August 2020, Version of Record 24 August 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107608