Applications of deep learning in high energy nuclear physics
Abstract
Deep learning methods are exquisitely tailored to uncover the structure in complex data and efficiently describe it with a finite number of parameters. This data-driven method has recently emerged in high-energy nuclear physics. Experimental observations of high-energy nuclear physics are mainly derived from relativistic heavy-ion collisions (HICs) at the small length high-energy scales, and lattice quantum chromodynamics (QCD) from the first-principle calculations, based on which the extraction of physical information or optimizing calculations has become the foundation of deep learning applications. Given this, this paper will first introduce the use of neural networks to extract the phase structure and key physical information in HICs and then focus on the combination of generative algorithms and lattice QCD calculations. Finally, a physics-driven deep learning method is presented, and the possible far-reaching impact of this new research paradigm on the field is discussed.
