Representation learning via serial robust autoencoder for domain adaptation

作者：

Highlights：

• A novel serial robust autoencoder is designed for domain adaptation.

• We design two novel autoencoder models and connect them in series to learn effective representations.

• We design an iterative approach to solve the proposed GRA model, and prove its convergence.

摘要

•A novel serial robust autoencoder is designed for domain adaptation.•We design two novel autoencoder models and connect them in series to learn effective representations.•We design an iterative approach to solve the proposed GRA model, and prove its convergence.

论文关键词：Domain adaptation,Serial autoencoder,Representation learning

论文评审过程：Received 10 April 2019, Revised 1 January 2020, Accepted 4 June 2020, Available online 14 June 2020, Version of Record 20 July 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.113635