TY - JOUR
T1 - Deterministic learning of hybrid Fuzzy Cognitive Maps and network reduction approaches
AU - Nápoles, Gonzalo
AU - Jastrzebska, Agnieszka
AU - Mosquera, Carlos
AU - Vanhoof, Koen
AU - Homenda, Wladyslaw
N1 - Funding Information:
The authors would like to thank the anonymous reviewers for their valuable and constructive feedback.
Publisher Copyright:
© 2020 Elsevier Ltd
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2020/4
Y1 - 2020/4
N2 - Hybrid artificial intelligence deals with the construction of intelligent systems by relying on both human knowledge and historical data records. In this paper, we approach this problem from a neural perspective, particularly when modeling and simulating dynamic systems. Firstly, we propose a Fuzzy Cognitive Map architecture in which experts are requested to define the interaction among the input neurons. As a second contribution, we introduce a fast and deterministic learning rule to compute the weights among input and output neurons. This parameterless learning method is based on the Moore-Penrose inverse and it can be performed in a single step. In addition, we discuss a model to determine the relevance of weights, which allows us to better understand the system. Last but not least, we introduce two calibration methods to adjust the model after the removal of potentially superfluous weights. (C) 2020 Elsevier Ltd. All rights reserved.
AB - Hybrid artificial intelligence deals with the construction of intelligent systems by relying on both human knowledge and historical data records. In this paper, we approach this problem from a neural perspective, particularly when modeling and simulating dynamic systems. Firstly, we propose a Fuzzy Cognitive Map architecture in which experts are requested to define the interaction among the input neurons. As a second contribution, we introduce a fast and deterministic learning rule to compute the weights among input and output neurons. This parameterless learning method is based on the Moore-Penrose inverse and it can be performed in a single step. In addition, we discuss a model to determine the relevance of weights, which allows us to better understand the system. Last but not least, we introduce two calibration methods to adjust the model after the removal of potentially superfluous weights. (C) 2020 Elsevier Ltd. All rights reserved.
KW - Fuzzy cognitive maps
KW - Hybrid models
KW - Inverse learning
KW - Interpretability
UR - http://www.scopus.com/inward/record.url?scp=85078744842&partnerID=8YFLogxK
U2 - 10.1016/j.neunet.2020.01.019
DO - 10.1016/j.neunet.2020.01.019
M3 - Article
SN - 0893-6080
VL - 124
SP - 258
EP - 268
JO - Neural Networks: The official journal of the International Neural Network Society, European Neural Network Society, Japanese Neural Network Society
JF - Neural Networks: The official journal of the International Neural Network Society, European Neural Network Society, Japanese Neural Network Society
ER -