A COGNITIVE APPROACH TO NEURAL NETWORK MODEL BASED ON THE COMMUNICATION SYSTEM BY AN INFORMATION CRITERIA

Authors : Alla HORİUSHKİNA, Serhat SEKER, Ufuk KORKMAZ

Pages : 1-4

View : 14 | Download : 9

Publication Date : 2018-12-01

Article Type : Research Paper

Abstract :In this paper presents the structural analogy between the artificial neural net insert ignore into journalissuearticles values(ANN); architecture and communication system principles. Thus, a feed forward auto-associative ANN is assumed like a communication system and the channel capacity is defined as an information measure related to the number of hidden nodes which are in one hidden layer of the neural net. Here, the channel capacity formula which is given for the communication channel is represented as information a criterion that is connected with Shannon’s entropy. Hence it is defined as information capacity under the interpretation of cognitive capacity and set theory approach is also used to define the ANN in manner of cognitive system.
Keywords : Neural Nets, Communication Channel, Learning System, Shannons Entropy, Cognitive Systems