This study presents an autoencoder-embedded opti-mization (AEO) algorithm which involves a bi-population coop-erative strategy for medium-scale expensive problems (MEPs). A huge search space can be compressed to an informative low-dimensional space by using an autoencoder as a dimension reduction tool. The search operation conducted in this low space facilitates the population with fast convergence towards the optima. To strike the balance between exploration and exploita-tion during optimization, two phases of a tailored teaching-learn-ing-based optimization (TTLBO) are adopted to coevolve solu-tions in a distributed fashion, wherein one is assisted by an autoencoder and the other undergoes a regular evolutionary pro-cess. Also, a dynamic size adjustment scheme according to prob-lem dimension and evolutionary progress is proposed to promote information exchange between these two phases and accelerate evolutionary convergence speed. The proposed algorithm is vali-dated by testing benchmark functions with dimensions varying from 50 to 200. As indicated in our experiments, TTLBO is suit-able for dealing with medium-scale problems and thus incorpo-rated into the AEO framework as a base optimizer. Compared with the state-of-the-art algorithms for MEPs, AEO shows extraordinarily high efficiency for these challenging problems, thus opening new directions for various evolutionary algorithms under AEO to tackle MEPs and greatly advancing the field of medium-scale computationally expensive optimization.

Meiji Cui;Li Li;MengChu Zhou;Jiankai Li;Abdullah Abusorrah;Khaled Sedraoui

School of Intelligent Manufacturing,Nanjing University of Science and Technology,Nanjing 210094,China;Department of Electronics and Information Engineering,Tongji University,Shanghai 201804,China;Helen and John C.Hartmann Department of Electrical and Computer Engineering,New Jersey Institute of Technology,Newark,NJ 07102 USA;Center of Research Excellence in Renewable Energy and Power Systems,Department of Electrical and Computer Engineering,Faculty of Engineering,and K.A.CARE Energy Research and Innovation Center,King Abdulaziz University,Jeddah 21589,Saudi Arabia

自动化学报（英文版）

[1]Meiji Cui;Li Li;MengChu Zhou;Jiankai Li;Abdullah Abusorrah;Khaled Sedraoui-.A Bi-population Cooperative Optimization Algorithm Assisted by an Autoencoder for Medium-scale Expensive Problems)[J].自动化学报（英文版）,2022(11):1952-1966

Expensive,TTLBO,coevolve

Bi,population,Cooperative,Optimization,Algorithm,Assisted,by,Autoencoder,Medium,scale,Problems,This,study,presents,autoencoder,embedded,AEO,which,involves,bi,coop,strategy,medium,expensive,problems,MEPs,huge,search,space,can,compressed,informative,low,dimensional,using,reduction,tool,operation,conducted,this,facilitates,fast,convergence,towards,optima,To,strike,balance,between,exploration,exploita,during,optimization,two,phases,tailored,teaching,learn,adopted,solu,distributed,fashion,wherein,one,assisted,other,undergoes,regular,evolutionary,cess,Also,dynamic,size,adjustment,scheme,according,progress,proposed,promote,information,exchange,these,accelerate,speed,vali,dated,testing,benchmark,functions,dimensions,varying,from,indicated,our,experiments,suit,able,dealing,thus,incorpo,rated,into,framework,optimizer,Compared,state,art,algorithms,shows,extraordinarily,high,efficiency,challenging,opening,new,directions,various,tackle,greatly,advancing,field,computationally

0.586607