Using multiobjective evolutionary algorithms to assess biological simulation models

被引:0
|
作者
Komuro, Rie [1 ]
Reynolds, Joel H. [2 ]
Ford, E. David [3 ]
机构
[1] Univ Auckland, Bioengn Inst, Level 6,70 Symonds St, Auckland, New Zealand
[2] US Fish & Wildlife Serv, Div Nat Resources, Anchorage, AK 99503 USA
[3] Univ Washington, Coll Forest Resources, Seattle, WA 98195 USA
基金
美国国家科学基金会; 美国国家环境保护局;
关键词
multiobjective optimization; Pareto frontier; binary discrepancy measures; process model; mechanistic model; model assessment; structural inference; elitism;
D O I
暂无
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
We introduce an important general Multiobjective Evolutionary Algorithm (MOEA) application - assessment of mechanistic simulation models in biology. These models are often developed to investigate the processes underlying biological phenomena. The proposed model structure must be assessed to reveal if it adequately describes the phenomenon. Objective functions are defined to measure how well the simulations reproduce specific phenomenon features. They may be continuous or binary-valued, e.g. constraints, depending on the quality and quantity of phenomenon data. Assessment requires estimating and exploring the model's Pareto frontier. To illustrate the problem, we assess a model of shoot growth in pine trees using an elitist MOEA based on Nondominated Sorting in Genetic Algorithms. The algorithm uses the partition induced on the parameter space by the binary-valued objectives. Repeating the assessment with tighter constraints revealed model structure improvements required for a more accurate simulation of the biological phenomenon.
引用
收藏
页码:560 / +
页数:3
相关论文
共 50 条
  • [1] Design of a TTC Antenna Using Simulation and Multiobjective Evolutionary Algorithms
    Moreno, Javier
    Gonzalez, Ivan
    Rodriguez, Daniel
    [J]. IEEE AEROSPACE AND ELECTRONIC SYSTEMS MAGAZINE, 2019, 34 (07) : 18 - 31
  • [2] Module identification from heterogeneous biological data using multiobjective evolutionary algorithms
    Calonder, Michael
    Bleuler, Stefan
    Zitzler, Eckart
    [J]. PARALLEL PROBLEM SOLVING FROM NATURE - PPSN IX, PROCEEDINGS, 2006, 4193 : 573 - 582
  • [3] Global Multiobjective Optimization Using Evolutionary Algorithms
    Thomas Hanne
    [J]. Journal of Heuristics, 2000, 6 : 347 - 360
  • [4] Multiobjective Groundwater Management Using Evolutionary Algorithms
    Siegfried, Tobias
    Bleuler, Stefan
    Laumanns, Marco
    Zitzler, Eckart
    Kinzelbach, Wolfgang
    [J]. IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2009, 13 (02) : 229 - 242
  • [5] MOLeCS: Using multiobjective evolutionary algorithms for learning
    Mansilla, EBI
    Guiu, JMGI
    [J]. EVOLUTIONARY MULTI-CRITERION OPTIMIZATION, PROCEEDINGS, 2001, 1993 : 696 - 710
  • [6] Global multiobjective optimization using evolutionary algorithms
    Hanne, T
    [J]. JOURNAL OF HEURISTICS, 2000, 6 (03) : 347 - 360
  • [7] Multiobjective optimisation of fuzzy controllers using evolutionary algorithms
    Klaassen, KP
    Litz, L
    [J]. UKACC INTERNATIONAL CONFERENCE ON CONTROL '98, VOLS I&II, 1998, : 1581 - 1586
  • [8] Speeding up backpropagation using multiobjective evolutionary algorithms
    Abbass, HA
    [J]. NEURAL COMPUTATION, 2003, 15 (11) : 2705 - 2726
  • [9] Multiobjective placement of electronic components using evolutionary algorithms
    Deb, K
    Jain, P
    Gupta, NK
    Maji, HK
    [J]. IEEE TRANSACTIONS ON COMPONENTS AND PACKAGING TECHNOLOGIES, 2004, 27 (03): : 480 - 492
  • [10] Multiobjective optimization using adaptive fuzzy/evolutionary algorithms
    Lee, MA
    Esbensen, H
    [J]. COMPUTERS AND THEIR APPLICATIONS - PROCEEDINGS OF THE ISCA 11TH INTERNATIONAL CONFERENCE, 1996, : 67 - 70