Non-exact complexity reduction of generalized neuro-fuzzy networks

In modern measurement, control, monitoring and fault diagnosis systems, there is an increasing need for the use of non-classical computing methods. On the other hand, in these systems the available time and resources are usually limited, so methods with lower computational complexity are needed. Thus, the need arises to have formal methods for the complexity reduction of different soft-computing techniques. This paper discusses a possible method for the non-exact reduction of generalized type neuro-fuzzy systems, and gives the necessary error-bounds of the reduction.