NEW DEVELOPED METHODS USED FOR DATA SCIENCE OPTIMIZATION AS STATE-OF-THE-ART

Abstract

Nowadays, there is a potential need in the new era of data science for introducing newly developed methods and algorithms to be used to formulate data science models by optimization solutions. We are very much concerned with studying and improving extraordinary work as Stat-of-the-Art methods and/or algorithms for solving data science problems with respect to its scalability, and efficiency; which mainly include gradient-descent based algorithms, derivative free algorithms. We do really believe that the best method which has the ability to meet our goals for optimization solutions; is to benefit from using machine learning capabilities. Optimization formulations and algorithms are both possible to lead the development of new optimization approaches that make significant changes presented by machine learning applications.