Epistemic Uncertainty Quantification in Physical Model using Machine Learning

Over the last few years, the field of Artifical Intelligence (AI) has undergone a remarkable improvement in many fields of application. However, some fields such as medics, safety, and finance are very sensitive and packed with uncertainties. Thus, the development of an AI system that able to deal with fundamental uncertainties which produces robust solutions has become more important than before. In this research, we propose a new paradigm of AI that would incorporate the system with a proper treatment of uncertainty that comes from our partial knowledge of the world (Epistemic) that results in a more robust and trustable AI by human beings.