Super-encoder with cooperative autoencoder networks

作者：

Highlights：

• A novel supervised and cooperative autoencoder network for dimensionality reduction.

• An autoencoder dealing with both latent code separability and re-constructability.

• Super-encoder is superior to existing dimensionality-reduction models.

• As a generative model, super-encoder is comparable with existing autoencoder models.

摘要

•A novel supervised and cooperative autoencoder network for dimensionality reduction.•An autoencoder dealing with both latent code separability and re-constructability.•Super-encoder is superior to existing dimensionality-reduction models.•As a generative model, super-encoder is comparable with existing autoencoder models.

论文关键词：Autoencoder,Dimensionality reduction,Feature extraction,Pattern recognition,Cooperative neural networks

论文评审过程：Received 14 May 2021, Revised 25 January 2022, Accepted 26 January 2022, Available online 1 February 2022, Version of Record 4 February 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.108562