Emotional Brain-Inspired Adaptive Fuzzy Decayed Learning for online prediction problems

Abstract

In this paper, we propose a Fuzzy Adaptive BrainInspired Emotional Decayed Learning named Fuzzy ADBEL. Fuzzy ADBEL is a computational model that models the forgetting process and inhibitory mechanism of the emotional brain. In the model, the fuzzy decay rate simulates the forgetting process, and the stimulus and learning weights are considered as fuzzy variables trained by fuzzy learning rules. The final output of the model is evaluated by a fuzzy decision making layer that simulates the inhibitory mechanism. The proposed Fuzzy ADBEL is utilized to predict the Kp, AE and Dst indices showing opposite behaviors and characterizing the chaotic activity of the earth's magnetosphere. Experimental results show that fuzzy approaches including Fuzzy ADBEL and ANFIS (Adaptive NeuroFuzzy Inference System) reaches steady state faster than non-fuzzy approaches, ADBEL and MLP (Multilayer Perceptron). Hence, we hope the proposed model can be used in real time chaotic time series prediction.

Brain Emotional Learning Based Intelligent Decoupler for Nonlinear Multi‐Input Multi‐Output Distillation Columns

“…Since then, BELBIC is increasingly being utilized in many control engineering applications such as electric motors [13,14], servo systems [15,16], motion control [17,18], and power systems [19,20]. Some recent researches utilize other intelligent techniques in cooperation with BELBIC to control the system such as fuzzy logic [21,22]. These applications had utilized BELBIC as a SISO system.…”

Type-2 Fuzzy Hybrid Controller Network for Robotic Systems

“…A typical BELC network consists of a sensory subsystem and a neural-network judgment subsystem [21]. The network judgment subsystem indirectly impacts the outputs of the sensory subsystem based on the input values [27], [28]. The inputs of the two subsystems are mapped from the network inputs by a receptive-field mechanism inspired from the CMAC network.…”

Online and stable parameter estimation based on normalized brain emotional learning model (NBELM)

“…The BELBIC controller has been used for an array of engineering systems such as launch vehicles, 20 nonlinear systems, 21-23 path tracking problems, 24,25 unmanned air vehicle controls, 26 the process controlling in the distillation column process, 27,28 and electromechanical brake systems. In the work of Lotfi and Akbarzadeh-T, 31 they have solved a multiple-input-multiple-output classification problem by pattern recognition, whereas in the other work of Lotfi and Akbarzadeh-T, 32 fuzzy learning rules are utilized in order to train learning weights for prediction of the activity indices. Lotfi et al 30 have proposed an algorithm for prediction of activity indices of the Earth's magnetosphere in a supervised manner, considering a decay rate in Amygdala learning rule.…”

“…Lotfi et al 30 have proposed an algorithm for prediction of activity indices of the Earth's magnetosphere in a supervised manner, considering a decay rate in Amygdala learning rule. In the work of Lotfi and Akbarzadeh-T, 31 they have solved a multiple-input-multiple-output classification problem by pattern recognition, whereas in the other work of Lotfi and Akbarzadeh-T, 32 fuzzy learning rules are utilized in order to train learning weights for prediction of the activity indices. It is shown that faster training, higher accuracy, and lower spatial complexity are the advantages of the BELM.…”