A New Classification Arithmetic for Multi-Image Classification in Genetic Programming

New classification algorithm for multi-image classification in genetic programming (GP) is introduced, which includes algorithm of centered dynamic range selection with the power value of arithmetic progression and that of centered dynamic range selection with quick-decreasing power value of arithmetic progression. In the training of the operating agent for GP multi-image classification, different sets of power values are tested to achieve a more suitable range of margin values for the improvement of the accuracy in classification. The tests show that the two approaches can be used for more accurate classification for complicated problems of multi-image classification, and work better than the algorithm of either static or dynamic range selection.