The projection adaptive natural gradient online algorithm for SVM

The training of Support Vector Machine (SVM) is an optimization problem of quadratic programming which can not be applied to the online training in real time applications or time-variant data source. The online algorithms proposed by other researchers have high computational complexity and slow training speed. In this paper the projection gradient and adaptive natural gradient is combined. The projection adaptive natural gradient online algorithm is proposed. The learning performance is compared via prediction of the concentration of component A of Continuous Stirred Tank Reactor. The results of simulation demonstrate that the time taken by the projection adaptive natural gradient online algorithm is much less than that of incremental algorithm, while keep similar prediction precision.