SIMULATION OF NEURAL NETWORK IN ALFVEN-SPIN WAVE PROPAGATION THROUGH FERROMAGNETIC SEMICONDUCTOR WITH DEEP LEARNING

Abstract

The expansion of artificial intelligence (AI) and machine learning (ML) has upped the standard for scientific studies and research, including physics. the development of dynamic graphic models of the physical processes being studied using machine learning and neural network algorithms, which convert theoretical concepts and abstract equations into compelling visual depictions of these phenomena. Purpose of this work is to show in what way computational representations can be used for simulation to analyze the effects of changing number densities of electrons and holes on the Alfven-spin wave propagation through ferromagnetic semiconductor with deep learning. Propagation region, non-propagation regions, cutoff regions and stop bands can all be observed and analyzed perfectly with this tool, in a much easier way. This simulation used machine learning (ML) and neural networks to route huge quantities of data and precisely crack multifaceted, theoretically stated calculations that were previously thought to be beyond computing via assimilating mathematical datasets and physical restraints into a Python-based structure.