Building Interpretable Systems in Real Time
J. Victor Ramos, Carlos Pereira, Antonio Carlos Dourado · 2010
Building interpretable learning machines from data, incrementally, with the capability for a-posteriori knowledge extraction, is still a big challenge. It is difficult to avoid some degree of redundancy and unnecessary complexity in the obtained models. This chapter presents two alternative ways for building interpretable systems, using fuzzy models or kernel machines. The goal is to control the complexity of the machine, either the number of rules or the number of kernels, using incremental learning techniques and merging of fuzzy sets. In the first case, online implementation of mechanisms for merging membership functions and rule base simplification, toward evolving first-order Takagi-Sugeno fuzzy systems (eTS), is proposed in order to improve the interpretability of the fuzzy models. As a comparative solution, kernel-based methods may also contribute to the interpretability issue, particularly the incremental learning algorithms. The interpretability is analyzed based on equivalence principles between local model networks and Takagi-Sugeno fuzzy inference systems. Controlled Vocabulary Terms fuzzy systems; knowledge based systems; learning systems; operating system kernels