A computationally fast Interval Type-2 Neuro-Fuzzy Inference System and its Meta-Cognitive projection based learning algorithm

摘要: In this paper, a computationally efficient Interval Type-2 Neuro-Fuzzy Inference System (IT2FIS) and its Meta-Cognitive projection based learning (PBL) algorithm is presented, together referred as PBL-McIT2FIS. A six layered network with cheap type-reduction technique proposed, rendering the inference mechanism faster. During learning, assumes that IT2FIS has no rules in beginning, adds to updates it depending on prediction error relative knowledge present current sample. As each sample presented network, meta-cognitive component of decides what-to-learn, when-to-learn how-to-learn it, instantaneous spherical potential Whenever new rule added or an existing updated, computes optimal output weights by minimizing total manner. The performance PBL-McIT2FIS evaluated set benchmark problem compared other state-of-the-art algorithms available literature. results indicate superior