Material properties share an intrinsic relationship with their structural attributes,making inverse design approaches crucial for discovering new materials with desired functionalities.Reinforcement Learning(RL)approa...Material properties share an intrinsic relationship with their structural attributes,making inverse design approaches crucial for discovering new materials with desired functionalities.Reinforcement Learning(RL)approaches are emerging as powerful inverse design tools,often functioning in discrete action spaces.This constrains their application in materials design problems,which involve continuous search spaces.Here,we introduce an RL-based framework CASTING(Continuous Action Space Tree Search for inverse design),that employs a decision tree-based Monte Carlo Tree Search(MCTS)algorithm with continuous space adaptation through modified policies and sampling.Using representative examples like Silver(Ag)for metals,Carbon(C)for covalent systems,and multicomponent systems such as graphane,boron nitride,and complex correlated oxides,we showcase its accuracy,convergence speed,and scalability in materials discovery and design.Furthermore,with the inverse design of super-hard Carbon phases,we demonstrate CASTING’s utility in discovering metastable phases tailored to user-defined target properties and preferences.展开更多
基金This work performed at the Center for Nanoscale Materials,a U.S.Department of Energy Office of Science User Facility,was supported by the U.S.DOE,Office of Basic Energy Sciences,under Contract No.DE-AC02-06CH11357This material is based on work supported by the DOE,Office of Science,BES Data,Artificial Intelligence,and Machine Learning at DOE Scientific User Facilities program(ML-Exchange).S.K.R.S.would also like to acknowledge the support from the UIC faculty start-up fund.This research used resources of the National Energy Research Scientific Computing Center(NERSC),a US Department of Energy Office of Science User Facility located at Lawrence Berkeley National Laboratory,operated under Contract No.DE-AC02-05CH11231The authors(SKRS and TDL)would like to acknowledge the Air Force Office of Scientific Research(AFOSR)for funding this research under Award#FA9550-20-1-0332,with Dr.Chipping Li as the program manager.
文摘Material properties share an intrinsic relationship with their structural attributes,making inverse design approaches crucial for discovering new materials with desired functionalities.Reinforcement Learning(RL)approaches are emerging as powerful inverse design tools,often functioning in discrete action spaces.This constrains their application in materials design problems,which involve continuous search spaces.Here,we introduce an RL-based framework CASTING(Continuous Action Space Tree Search for inverse design),that employs a decision tree-based Monte Carlo Tree Search(MCTS)algorithm with continuous space adaptation through modified policies and sampling.Using representative examples like Silver(Ag)for metals,Carbon(C)for covalent systems,and multicomponent systems such as graphane,boron nitride,and complex correlated oxides,we showcase its accuracy,convergence speed,and scalability in materials discovery and design.Furthermore,with the inverse design of super-hard Carbon phases,we demonstrate CASTING’s utility in discovering metastable phases tailored to user-defined target properties and preferences.