Prediction of Stochastic Cognitive Neural Schema with neural network paradigm of latent semantic nodes in Autobot - Humanoid robot using non linear regression of Gaussian sigmoidal curves in Boltzmann Normalisation

• Published in: International Information Institute (Tokyo). Information Vol. 17; no. 8; p. 3971
• Main Authors: Ramadoss, Ashok Kumar, Krishnaswamy, Marimuthu
• Format: Journal Article
• Language: English
• Published: Koganei International Information Institute 01.08.2014
• Subjects: Emotions; Gaussian; Humanoid; Learning; Neural networks; Regression; Regression analysis; Robots; Semantics; Tasks
• ISSN: 1343-4500; 1344-8994

The Intelligent system is organized and implemented in a semantic fashion in neural networks according to the local specialization of problem solving where a semantic of neural networks implements an inter-related group of knowledge, such as a Neural Schema but differ in their input and output patterns in the hidden and output layer with passive nodes, which enables the behaviour of the Artificial Brain through the Stimulus and response learning in this Intelligent systems. A neural network model can be organised just as a schema structure, and experiments were conducted to validate the derived cognitive knowledge structure in neural networks. The experiments were based on a task of semantic model as found insights of the model of a neural learning Schema, using non linear regression and Gaussian curves of Boltzmann Normalisation and COX model.