A Review of Perspectives on Establishing the Cost Function, Regularization and Optimization for Machine Learning Techniques Along with Applications

Author(s):

Advait Pravin Savant , Sardar Patel Institute of Technology

Keywords:

Machine Learning, Cost Function, Regularization, Entropy, Gradient Descent

Abstract:

Philosophers across ages have wondered if a machine could perform tasks that require human cognitive abilities. With the formalization of computational theory and the advent of electronic computers, scientists have worked on developing computational models for intelligence and learning. Machine learning is the field of AI that provides computer systems the ability to improve their efficiency at performing a certain task as measured by a performance measure based on the experience that is given. With the increase in processing power of modern computers along with telecommunication technology, there is large amount of data that is available which enables us to build models of complex systems and extract information from the system, create representations of knowledge, recognize patterns from data and improve our ability to solve complex problems. From astronomical data analysis to credit card fraud detection, from medical data analysis to stock market predictions, today machine learning is everywhere. In this paper I attempt to review, describe and compare certain machine learning paradigms.

Other Details:

Manuscript Id	:	IJSTEV6I6001
Published in	:	Volume : 6, Issue : 6
Publication Date	:	01/01/2020
Page(s)	:	1-6

Download Article