Testing Machine Learning at Classical Electrodynamics

• Autor: Nieto-Chaupis, Huber
• Formato: artículo
• Fecha de Publicación: 2021
• Institución: Universidad Autónoma del Perú
• Repositorio: AUTONOMA-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorio.autonoma.edu.pe:20.500.13067/1647
• Enlace del recurso: https://hdl.handle.net/20.500.13067/1647; https://doi.org/10.23919/SpliTech52315.2021.9566432
• Nivel de acceso: acceso restringido
• Materia: Electrodynamics; Atmospheric measurements; Volume measurement; Machine learning; Tools; Particle measurements; Minimization; https://purl.org/pe-repo/ocde/ford#2.02.04

Like physics or another laws-based basic science, machine learning might also be a firm methodology to solve physics problems by the which a kind of optimization and minimization of energy are needed. Expressed at the Mitchell's principles, machine learning can be seen as a strategy that allows to improve physical actions such as observation and measurement. In the classical territory, one can project the well-known electrodynamics over the steps: (i) task, (ii) performance, and (iii) experience. With this one might to guarantee a kind of learning to face a next similar situation and so on. This paper try to solve the problem of a charged particle inside a cylindrical volume but emphasizing its energy and its measurement. Simulations have shown that machine learning can also be an alternative tool to solve physics problems that require of minimization of energy.