A novel deep auto-encoder considering energy and label constraints for categorization

作者：

Highlights：

• ELDAE is proposed by integrating energy and label constraints.

• The energy constraint improves the probability of network for fitting data.

• The label constraint is integrated to describe categorization rule.

• The performance is exhibited with the higher classification accuracy.

摘要

•ELDAE is proposed by integrating energy and label constraints.•The energy constraint improves the probability of network for fitting data.•The label constraint is integrated to describe categorization rule.•The performance is exhibited with the higher classification accuracy.

论文关键词：Deep learning,Deep auto-encoder,Semi-supervised learning,Energy constraint,Categorization

论文评审过程：Received 9 September 2019, Revised 19 February 2021, Accepted 21 March 2021, Available online 26 March 2021, Version of Record 4 April 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114936