Journal
Scientific and technical journal of information technologies, mechanics and optics
UDK004.4’242
Issue:8 (53)
Download PDF0 Kbyte
A machine learning method for complex systems with many input variables and many output actions is described in this paper. This method allows taking into consideration behaviour restrictions. It greatly reduces size of a search space. This method also allows splitting system into subsystems on the base of subproblems. Solutions for subproblems can be united with rather complex conditions on transitions between them.