A proposal on centralised and distributed optimisation via proportional–integral–derivative controllers(PID)control perspective

作     者：Jiaxu Liu Song Chen Shengze Cai Chao Xu 

作者机构：School of Mathematical SciencesZhejiang UniversityHangzhouChina The State Key Laboratory of Industrial Control TechnologyInstitute of Cyber-Systems and ControlZhejiang UniversityHangzhouChina 

出 版 物：《IET Cyber-Systems and Robotics》 (智能系统与机器人（英文）)

年 卷 期：2023年第5卷第4期

页      面：40-48页

核心收录：

学科分类：08[工学] 080401[工学-精密仪器及机械] 0804[工学-仪器科学与技术] 080402[工学-测试计量技术及仪器] 

基　　金：National Key Research and Development Program of China,Grant/Award Number:2019YFB1705800 National Natural Science Foundation of China,Grant/Award Number:61973270 Science and Technology Innovation 2030 New Generation Artificial Intelligence Major Project,Grant/Award Number:2018AAA0100902 

主　　题：control deep learning deep neural network machine learning 

摘      要：Motivated by the excellent performance of proportional–integral–derivative controllers(PIDs)in the field of control,the authors injected the philosophy of PID into optimi-sation and introduced two types of novel PID optimisers from a continuous-time view,which benefit from the idea that discrete-time optimisation algorithm can be modelled as a continuous dynamical system/controlled *** centralised optimisation,the au-thors discuss the idea of the first-order PID optimiser and the second-order accelerated PID ***,this framework is extended into distributed optimisation settings,and a distributed PID optimiser is ***,some numerical examples are given to verify our ideas.