Image Super-resolution Inspired Electron Density Prediction
This work addresses the challenge of efficient and accurate electron density prediction in quantum mechanics, offering a novel method that leverages image processing techniques for broader applicability in computational chemistry.
The paper tackled the problem of predicting accurate ground-state electron densities for molecules by treating them as 3D grayscale images and using a convolutional residual network to refine crude initial guesses, resulting in a model that outperforms all prior approaches and is applicable to unseen conformations and elements.
Drawing inspiration from the domain of image super-resolution, we view the electron density as a 3D grayscale image and use a convolutional residual network to transform a crude and trivially generated guess of the molecular density into an accurate ground-state quantum mechanical density. We find that this model outperforms all prior density prediction approaches. Because the input is itself a real-space density, the predictions are equivariant to molecular symmetry transformations even though the model is not constructed to be. Due to its simplicity, the model is directly applicable to unseen molecular conformations and chemical elements. We show that fine-tuning on limited new data provides high accuracy even in challenging cases of exotic elements and charge states. Our work suggests new routes to learning real-space physical quantities drawing from the established ideas of image processing.