Expert system gradient descent style training: Development of a defensible artificial intelligence technique

作者：

Highlights：

• Presents new artificial intelligence technique based on using machine learning principles with a base expert system.

• Describes how this can mitigate bias and other issues caused by the use of neural networks.

• Characterizes system efficacy and performance.

摘要

•Presents new artificial intelligence technique based on using machine learning principles with a base expert system.•Describes how this can mitigate bias and other issues caused by the use of neural networks.•Characterizes system efficacy and performance.

论文关键词：Expert systems,Gradient descent,Defensible artificial intelligence,Machine learning,Training

论文评审过程：Received 15 March 2021, Revised 8 May 2021, Accepted 30 June 2021, Available online 2 July 2021, Version of Record 7 July 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.107275