Automatic design of quantum feature maps

We propose a new technique for the automatic generation of optimal ad-hoc ans\"atze classification by using quantum support vector machine (QSVM). This efficient method is based on NSGA-II multiobjective genetic algorithms which allow both maximize accuracy and minimize ansatz size. It demonstrated validity practical example with non-linear dataset, interpreting resulting circuit its outputs. also show other application fields that reinforce method, comparison classical classifiers in order to understand advantages learning.