A new neural network approach improves how we model quantum matter
In short: Researchers have developed a new way to describe the behavior of electrons in matter by training a neural network on complex molecular data. By using both real-world molecules and artificial systems with fractional charges, the team created a functional called DM21. This tool more accurately predicts how electrons interact in difficult scenarios, such as when they are strongly correlated or spread out across different atoms.
Current methods for modeling quantum matter often fail because they violate certain mathematical rules. This study addresses those flaws by using a data-driven approach that incorporates specific physical constraints. While the results show significant improvements over traditional models for various chemical structures, the study's reliance on training data means its accuracy is tied to the quality and variety of the systems it learns from.
Why it matters: This research provides a data-driven pathway toward creating a more perfect mathematical description of how matter behaves at the quantum level.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
- Source: Research: Kirkpatrick et al. (2021). Pushing the frontiers of density functionals by solving the fractional electron problem. Science. · cc-by
To cite this card, link to https://intellectplus.co.uk/app/science-research-a-new-neural-network-approach-improves-how-we-model-quantum.
How we source, check and correct cards: Editorial standards.
General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.
Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?
Related
More in Science · Latest · A–Z · All topics