Automated Machine Learning (AutoML): The Future of Computational Intelligence

Abstract

AbstractComputer science controls every task in today’s environment, and everything in the sector attempts to automate the task. The basic essence of computer science is to minimize human effort and the core of reducing human involvement in any task is to automate. Machine learning (ML) has become an important part of many aspects of our daily life. High-performance machine learning applications, on the other hand, necessitate the use of highly qualified data scientists and domain specialists. Automated machine learning (AutoML) aims to reduce the need for data scientists by allowing domain experts to automatically construct machine learning applications without extensive statistical and machine learning knowledge. This paper provides an overview of existing AutoML methodologies as well as information on related technologies. We present and investigate important AutoML techniques and methodologies along with present challenges and future research directions. We also analyze various security threats that can be posed to the machine learning models and AutoML.KeywordsAutoMLMachine learningAutomationSecuritySecurity threats