Display Accessibility Tools

Accessibility Tools


Highlight Links

Change Contrast

Increase Text Size

Increase Letter Spacing

Dyslexia Friendly Font

Increase Cursor Size

Mauwiz Chaudhary

Electronic structure problems are key to many insights in a wide variety of scientific fields, including chemistry, materials science, and drug discovery. Current ab initio methods, such as density functional theory (DFT), for calculating the ground state electronic density of a system and its corresponding energy are computationally expensive. To reduce the computational costs, machine learning techniques have recently been applied to compute the ground state energy and in a few cases, the electronic density as well. In this work we introduce a technique for calculating the ground state electronic density at a reduced computational cost relative to DFT. Our method utilizes tools from optimal transport and uses a machine learning model reminiscent of a conditional adversarial network with a U-net architecture.