Evolutionary Optimization Methods for High-Dimensional Expensive Problems:A Survey

Evolutionary computation is a rapidly evolving field and the related algorithms have been successfully used to solve various real-world optimization problems.The past decade has also witnessed their fast progress to solve a class of challenging optimization problems called high-dimensional expensive problems(HEPs).The evaluation of their objective fitness requires expensive resource due to their use of time-consuming physical experiments or computer simulations.Moreover,it is hard to traverse the huge search space within reasonable resource as problem dimension increases.Traditional evolutionary algorithms(EAs)tend to fail to solve HEPs competently because they need to conduct many such expensive evaluations before achieving satisfactory results.To reduce such evaluations,many novel surrogate-assisted algorithms emerge to cope with HEPs in recent years.Yet there lacks a thorough review of the state of the art in this specific and important area.This paper provides a comprehensive survey of these evolutionary algorithms for HEPs.We start with a brief introduction to the research status and the basic concepts of HEPs.Then,we present surrogate-assisted evolutionary algorithms for HEPs from four main aspects.We also give comparative results of some representative algorithms and application examples.Finally,we indicate open challenges and several promising directions to advance the progress in evolutionary optimization algorithms for HEPs.

On the symplectic superposition method for free vibration of rectangular thin plates with mixed boundary constraints on an edge

A novel symplectic superposition method has been proposed and developed for plate and shell problems in recent years.The method has yielded many new analytic solutions due to its rigorousness.In this study,the first endeavor is made to further developed the symplectic superposition method for the free vibration of rectangular thin plates with mixed boundary constraints on an edge.The Hamiltonian system-based governing equation is first introduced such that the mathematical techniques in the symplectic space are applied.The solution procedure incorporates separation of variables,symplectic eigen solution and superposition.The analytic solution of an original problem is finally obtained by a set of equations via the equivalence to the superposition of some elaborated subproblems.The natural frequency and mode shape results for representative plates with both clamped and simply supported boundary constraints imposed on the same edge are reported for benchmark use.The present method can be extended to more challenging problems that cannot be solved by conventional analytic methods.

New analytic solutions to 2D transient heat conduction problems with/without heat sources in the symplectic space

The two-dimensional(2D)transient heat conduction problems with/without heat sources in a rectangular domain under different combinations of temperature and heat flux boundary conditions are studied by a novel symplectic superposition method(SSM).The solution process is within the Hamiltonian system framework such that the mathematical procedures in the symplectic space can be implemented,which provides an exceptional direct rigorous derivation without any assumptions or predetermination of the solution forms compared with the conventional inverse/semi-inverse methods.The distinctive advantage of the SSM offers an access to new analytic heat conduction solutions.The results obtained by the SSM agree well with those obtained from the finite element method(FEM),which confirms the accuracy of the SSM.

Deleterious variants in RNF111 impair female fertility and induce premature ovarian insufficiency in humans and mice

Premature ovarian insufficiency(POI)is a heterogeneous female disorder characterized by the loss of ovarian function before the age of 40.It represents a significant detriment to female fertility.However,the known POI-causative genes currently account for only a fraction of cases.To elucidate the genetic factors underlying POI,we conducted whole-exome sequencing on a family with three fertile POI patients and identified a deleterious missense variant in RNF111.In a subsequent replication study involving 1,030 POI patients,this variant was not only confirmed but also accompanied by the discovery of three additional predicted deleterious RNF111 variants.These variants collectively account for eight cases,representing 0.78%of the study cohort.A further study involving 500 patients with diminished ovarian reserve also identified two additional RNF111 variants.Notably,RNF111 encodes an E3 ubiquitin ligase with a regulatory role in the TGF-β/BMP signaling pathway.Our analysis revealed that RNF111/RNF111 is predominantly expressed in the oocytes of mice,monkeys,and humans.To further investigate the functional implications of RNF111 variants,we generated two mouse models:one with a heterozygous missense mutation(Rnf111+/M)and another with a heterozygous null mutation(Rnf111^(+/-)).Both mouse models exhibited impaired female fertility,characterized by reduced litter sizes and small ovarian reserve.Additionally,RNA-seq and quantitative proteomics analysis unveiled that Rnf111 haploinsufficiency led to dysregulation in female gonad development and negative regulation of the BMP signaling pathway within mouse ovaries.In conclusion,our findings strongly suggest that monoallelic deleterious variants in RNF111 can impair female fertility and induce POI in both humans and mice.