Regensburg 2022 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

QI: Fachverband Quanteninformation

QI 6: Quantum Information: Concepts and Methods

QI 6.4: Talk

Tuesday, September 6, 2022, 10:30–10:45, H9

Quantum Convolutional Neural Network as a Phase Detection Circuit on the Toric Code — •Leon Sander, Nathan McMahon, and Michael Hartmann — Chair of Theoretical Physics, Friedrich-Alexander-Universität Erlangen Nürnberg, Germany

Understanding macroscopic behaviour of quantum materials is an interesting challenge in the field of quantum technologies. This macroscopic behaviour can be evaluated by the examination of quantum phases. Consequently, recognising the phase of a given input state is an important problem, which is often solved by measuring the corresponding order parameter. However, previous work by Cong et al. and Hermann et al. suggests quantum convolutional neural networks (QCNN) are an alternative method of phase detection that can also improve sampling efficiency near the phase boundary compared to direct measurements.

We construct a QCNN designed to act as a phase recognition circuit that determines whether certain magnetic/Ising type perturbations are sufficient to induce a phase transition in the toric code. The choice to study this quantum error correcting code can be motivated as it promises to reveal connections between quantum information and quantum phase transitions.