Beyond Jacob’s Ladder: A Physics-driven AI Approach to Density Functional Theory
Abstract
Abstract Density Functional Theory (DFT) has become a widely used approach for accurate prediction of electronic structure and thermodynamic properties of molecular systems - from single molecules to large-scale systems. Although the accuracy of DFT functionals has significantly improved over the last decade, this is typically achieved through inclusion of more complexity in the functional’s description and larger basis sets. This significantly prevents applications of DFT to realistic molecular systems beyond single medium-sized molecules. The recent development of AlphaFold has demonstrated AI algorithms’ ability to learn physics, whereby achieving tremendous predictive power for the structure of peptides. Here we show that the inclusion of a more complex formulation in DFT can be overcome by training a machine learning (ML) code, termed ML-DFT, to extract the needed information from molecule’s electron density using a neural network, whereby robustly learning essential physics governing the behaviour of molecular systems. The ML code was trained to predict atomisation energies of molecules of varying functionality and intermolecular interaction energies of diverse neutral and charged molecular complexes, producing accuracy below of 1 kcal/mol and outperforming the best available DFT functionals. This remarkable outcome was achieved by constructing the loss function based on relative electronic energies to directly compare with thermodynamic properties – atomisation…
Read the paper
Related studies
- Preliminary Toxicity Analysis of 3DCRT versus IMRT on the High Dose Arm of the RTOG 0126 Prostate Cancer Trial
- Adaptive Radiotherapy for Head and Neck Cancer: Initial Clinical Outcomes from a Prospective Trial
- American Association of Physicists in Medicine Task Group 263: Standardizing Nomenclatures in Radiation Oncology
- Motion Management Strategies and Technical Issues Associated with Stereotactic Body Radiotherapy of Thoracic and Upper Abdominal Tumors: A Review from NRG Oncology
- High-dose-rate interstitial brachytherapy as a monotherapy for localized prostate cancer: treatment description and preliminary results of a phase I/II clinical trial
- Highly efficient training, refinement, and validation of a knowledge-based plan quality control system for radiotherapy clinical trials
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.