Communications of the ACM
ML Tool Sorts the Nuances of Quantum Data
An interdisciplinary team of researchers from Cornell and Harvard universies developed a machine learning tool to parse quantum matter and make crucial distinctions in the data.
Credit: IBM.
A machine learning tool developed by researchers at Cornell University and Harvard University can parse quantum matter and identify relationships among microscopic properties in the data.
The Cornell researchers created Correlation Convolutional Neural Networks (CCNN), an "interpretable architecture" that can determine the most important correlations in image-like data.
The Harvard researchers tested CCNN by simulating a fermionic Hubbard model using quantum gas microscopy. They generated synthetic data for geometric string theory and pi-flux theory, which are difficult to distinguish.
However, CCNN distinguished between the two simulations by identifying correlations in the data to the fourth order.
Cornell's Eun-Ah Kim said, "We're forcing the neural network to choose one or two features that help it the most in coming up with the right assessment. And by doing so we can figure out what are the critical aspects, the core essence, of what defines a state or phase."
From Cornell University Chronicle
View Full Article
Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA
No entries found