内嵌基因表达式编程及其在函数发现中的应用  

Embedded Gene Expression Programming and Its Application in Function Mining

作　　者：向勇[1] 唐常杰[2] 朱明放[3] 陈瑜[2] 代术成[2] 

机构地区：[1]成都电子机械高等专科学校计算机工程系,成都610031 [2]四川大学计算机学院,成都610065 [3]江苏技术师范学院计算机工程学院,江苏常州213001

出　　处：《电子科技大学学报》2011年第1期116-121,共6页Journal of University of Electronic Science and Technology of China

基　　金：国家自然科学基金(60773169);国家"十一五"科技支撑计划(2006BAI05A01);四川省教育厅科研资助(2006B067)

摘　　要：为了提高表达效率,提出了新的基因解码方案,形成了内嵌基因表达式编程算法EGEP;提出了极大表达树、嵌套表达树和拼接表达树等概念;分析了基因的表达空间和算法的复杂度。实验表明,该算法提高了函数发现的成功率;在小规模种群的函数中其能力明显优于GEP。在单基因情况下,目标为一元函数和二元函数时,EGEP平均成功辈数分别为GEP算法的25.5%和16.3%;各种规模下,在EGEP算法中二元函数的成功率平均比GEP提高43%以上。Gene Expression Programming is effective for function mining.In gene expression usually exist some un-expressed introns.To improve the expression efficiency,this paper makes following contributions： Proposed an evolutionary algorithm embedded gene expression programming（EGEP） based on a new decoding method of gene;Proposed some new concepts,i.e.the maximum expression tree,nested expression tree and spliced expression tree;Analyzed the expression space of gene and the complexity of algorithm.Extensive experiments show that the success rate is improved greatly and under the small size population,the ability of mining function surpasses GEP apparently.In single gene algorithms,when the objective functions are bivariate function and single-variable function,the ratios of the convergence generation of EGEP to that of GEP are 25.5% and 16.3% respectively;compared with GEP,the success rate of EGEP is averagly increased by 43% in bivariate function mining.

关 键 词：函数发现 遗传算法 基因表达式编程 基因内区 

分 类 号：TP311.6[自动化与计算机技术—计算机软件与理论]