End-to-end differentiability and tensor processing unit computing to accelerate materials’ inverse design  

作　　者：Han Liu Yuhan Liu Kevin Li Zhangji Zhao Samuel S.Schoenholz Ekin D.Cubuk Puneet Gupta Mathieu Bauchy 

机构地区：[1]SOlids inFormaTics AI-Laboratory(SOFT-AI-Lab),College of Polymer Science and Engineering,Sichuan University,Chengdu 610065,China [2]AIMSOLID Research,Wuhan 430223,China [3]Physics of AmoRphous and Inorganic Solids Laboratory(PARISlab),Department of Civil and Environmental Engineering,University of California,Los Angeles,CA 90095,USA [4]Department of Computer Science,University of California,Los Angeles,CA 90095,USA [5]Google Research,Brain Team,Mountain View,CA,USA [6]Department of Electrical and Computer Engineering,University of California,Los Angeles,CA 90095,USA

出　　处：《npj Computational Materials》2023年第1期1072-1083,共12页计算材料学（英文）

基　　金：H.L.acknowledges funding from the Fundamental Research Funds for the Central Universities under the Grant No.YJ202271;M.B.acknowledges the National Science Foundation under the Grant No.DMREF-1922167;TPU computing time was provided by a grant allocation from Google’s TensorFlow Research Cloud(TFRC)program.

摘　　要：Numerical simulations have revolutionized material design.However,although simulations excel at mapping an input material to its output property,their direct application to inverse design has traditionally been limited by their high computing cost and lack of differentiability.Here,taking the example of the inverse design of a porous matrix featuring targeted sorption isotherm,we introduce a computational inverse design framework that addresses these challenges,by programming differentiable simulation on TensorFlow platform that leverages automated end-to-end differentiation.Thanks to its differentiability,the simulation is used to directly train a deep generative model,which outputs an optimal porous matrix based on an arbitrary input sorption isotherm curve.Importantly,this inverse design pipeline leverages the power of tensor processing units(TPU)—an emerging family of dedicated chips,which,although they are specialized in deep learning,are flexible enough for intensive scientific simulations.This approach holds promise to accelerate inverse materials design.

关 键 词：ENOUGH inverse ISOTHERM 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]