abstract = "The canonical form of Genetic Programming (GP) relies
almost exclusively on the crossover operator for
exploring the solution space. But the crossover
operator makes individuals change randomly and
indirectly during the whole process, it not only give
the individuals the evolutionary chance but also cause
certain degeneracy. Here, based on the analysis of the
cross-operating characteristics and the study of
individual structure, a new form of vaccine and a new
autonomous obtaining method is proposed. The concept of
immunity is integrated into the GP and a novel
algorithm-immune GP is proposed. Experiments show that
the result obtained by the proposed algorithm restrains
the degenerative phenomena during evolution, so as to
improve the algorithm efficiency.",