Reducing the Cost of Quantum Chemical Data By Backpropagating Through Density Functional Theory

2024/02/06 by Alexander Mathiasen, Mathiasen, Alexander, Hatem Helal +15
Biochemistry, Genetics and Molecular Biology · Chemistry · #FOS: Computer and information sciences #History and advancements in chemistry #Machine Learning (cs.LG) #Metabolomics and Mass Spectrometry Studies #Various Chemistry Research Topics

paper · pdf · doi:10.48550/arxiv.2402.04030

openalex publication_date 2024/02/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Density Functional Theory (DFT) accurately predicts the quantum chemical properties of molecules, but scales as O(N_electrons³). Schütt et al. (2019) successfully approximate DFT 1000x faster with Neural Networks (NN). Arguably, the biggest problem one faces when scaling to larger molecules is the cost of DFT labels. For example, it took years to create the PCQ dataset (Nakata & Shimazaki, 2017) on which subsequent NNs are trained within a week. DFT labels molecules by minimizing energy E(⋅ ) as a "loss function." We bypass dataset creation by directly training NNs with E(⋅ ) as a loss function. For comparison, Schütt et al. (2019) spent 626 hours creating a dataset on which they trained their NN for 160h, for a total of 786h; our method achieves comparable performance within 31h.

Related

• Deep Neural Network Computes Electron Densities and Energies of a Large Set of Organic Molecules Faster than Density Functional Theory (DFT)
• Physical machine learning outperforms "human learning" in Quantum Chemistry
• A Density Functional Tight Binding Layer for Deep Learning of Chemical Hamiltonians
• Towards A Transferable Acceleration Method for Density Functional Theory
• Incorporating Scientific Knowledge into Neural Network Density Functionals