The foundation was founded by Thomas Sumner.
Physicists are trying to crack the code of quantum systems.
Predicting the properties of a molecule involves calculating the collective behavior of its electrons. Predicting properties such as superconductivity could one day help researchers develop new drugs. electrons can become entangled with one another, meaning they can't be treated individually Even the most powerful computers can't get to any system with more than a few particles.
The problem has been solved by quantum physicists at the cole Polytechnique Fédérale de Lausanne in Switzerland. Adding extra "ghost" electrons that interact with the system's actual electrons allowed them to create a way to mimic entanglement.
A neural network is used to control the behavior of the electrons. The network has to make changes until it finds a solution that can be projected back into the real world.
August 3 is when the physicists present their method.
The electrons can be treated as if they don't talk to each other, according to the study lead author. We're trying to describe the interactions between the particles that live in the physical system.
The physicists show in the new paper that their approach matches or surpasses competing methods.
As a test bed, we applied this to simple things, but now we are trying this on more realistic problems. If you have a good way of getting the wave functions of complex molecules, you can do a lot of different things.
The long-term goal is for researchers to be able to predict the properties of a molecule without having to make it in a lab. With a few clicks of a mouse, they might be able to test a bunch of different Molecules for a desired Pharmaceutical Property. It's a big deal to make big molecule simulations.
The paper was co-authored by Robledo and George.
The new work is an evolution of a paper written by Carleo and Troyer a year ago. The added particles weren't full-blown electrons. They had a single property called spin.
"When I was at the CCQ in New York, I was obsessed with the idea of finding a version of neural network that would describe the way electrons behave, and I really wanted to find a generalization of the approach we introduced back in 2017," Carleo said. We have found an elegant way of hiding hidden particles that are not spins.
More information: Javier Robledo Moreno et al, Fermionic wave functions from neural-network constrained hidden states, Proceedings of the National Academy of Sciences (2022). DOI: 10.1073/pnas.2122059119 Provided by Simons Foundation
All Rights Reserved © 2024
