Sign In

Communications of the ACM

ACM TechNews

AI Enables Efficiencies in Quantum Information Processing


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
The researchers implemented a system that reconstructed quantum states and standard, more computationally intensive methods.

A new machine learning framework could pave the way for small, mobile quantum networks.

Credit: Shutterstock

Researchers from the U.S. Army Combat Capabilities Development Command's Army Research Laboratory and Tulane University integrated machine learning with quantum information science, applying photon measurements to reassemble the quantum state of an unknown system.

The team used quantum state tomography to define an unknown quantum system, preparing and measuring identical unknown quantum systems and employing a complex computational process to ascertain the system most consistent with the measurement results.

A recent implementation reconstructed quantum states and standard, more computationally intensive methods—and outperformed those techniques in several instances.

The team modeled familiar sources of measurement error to train the machine learning artificial intelligence, then refined the system when measurements were noisy or entirely absent.

The actual reconstruction demands relatively modest resources, because the team can pile the expensive computation into the training process.

From U.S. Army Research Laboratory
View Full Article

 

Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA


 

No entries found