On Application of the Method of Potential Functions to the Stochastic Problem of the Optimal Partitioning of Sets
Elena Mikhailovna Kiseleva, Константин Александрович Кузнецов · Journal of Automation and Information Sciences · 1999
For the stochastic problem of optimal partitioning of sets (i.e. problems in which functions appearing in the partitioning performance functional and constraints depend on a random parameter), an algorithm of decision rule construction is proposed based on the solution of a finite sequence of determinate partitioning problems and on the use of the obtained partition as a teacher in the pattern recognition learning problem.