Deep Reinforcement Learning for Slow Diffusion Processes in Materials

We have applied Deep Reinforcement Learning to investigate slow processes in materials. Specifically, two systems are investigated: (1) diffusion in water in silica glass, (2) diffusion of molecular hydrogen in crystalline and amorphous polymers.

This research was done in collaboration with Ankit Mishra, Tian Sang, Rajiv K. Kalia, Aiichiro Nakano, and Priya Vashishta

This research was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Chemical Sciences, Geosciences, and Bioscience Division, Geosciences Program under Award DE-SC0025222.