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

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.

Automated summary

Numerical simulations have greatly impacted material design, but their application to inverse design has been limited due to high computing cost and lack of differentiability. This study introduces a computational framework that addresses these challenges by using differentiable simulations on the TensorFlow platform to train a deep generative model. The model outputs an optimal porous matrix based on any input sorption isotherm curve. The use of tensor processing units (TPUs) enhances the effectiveness of this inverse design approach.