Identification and removal of contaminants in sEMG recordings through a methodology based on Fuzzy Inference and Actor-Critic Reinforcement learning
作者：
Highlights：
• The proposed algorithm performs contaminant identification by unsupervised learning.
• The proposed algorithm enables the learning’s continuous adaptation.
• The proposed algorithm enables online training.
• Results of the unsupervised training are comparable than that of supervised methods.
• Handcrafted features are efficient to discriminate different contaminant types.
摘要
•The proposed algorithm performs contaminant identification by unsupervised learning.•The proposed algorithm enables the learning’s continuous adaptation.•The proposed algorithm enables online training.•Results of the unsupervised training are comparable than that of supervised methods.•Handcrafted features are efficient to discriminate different contaminant types.
论文关键词：Actor-Critic Reinforcement Learning,sEMG signal contamination,Electromyography,Fuzzy Inference System
论文评审过程：Received 29 October 2021, Revised 30 May 2022, Accepted 3 June 2022, Available online 9 June 2022, Version of Record 15 June 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.117772
