Parallel Gene Expression Programming Algorithm Based on Simulated Annealing Method
This paper presents a hybrid GEP parallel algorithm, whieh combines simulated annealing and genetic mechanism to avoid the problem of premature convergence. Meanwhile, it imports multi-population strategy and migration operator, and a coarsegrained parallel computing on finite number of CPUs is implemented. Experiments on sequence induction, function finding and sunspot prediction show that the algorithm has better search ability and higher stability than basic GEP.
gene expression programmingsimulate annealingmulti population strategyMPIparallel algorithmsymbolic regression
国家科技期刊平台
NSTL
万方数据
维普
