Implementation of Quantum Machine Learning for Electronic Structure Calculations of Periodic Systems on Quantum...

Area of Science:

• Quantum Computing
• Materials Science
• Computational Chemistry

Background:

• Quantum machine learning (QML) extends machine learning into quantum regimes, leveraging quantum properties for enhanced computational power.
• QML methods are increasingly used for solving quantum many-body systems, showing promise in electronic structure calculations for various material types.
• Hybrid approaches combining classical optimization with quantum algorithms offer efficiency and ease of implementation.

Purpose of the Study:

• To benchmark a hybrid quantum machine learning approach for electronic structure calculations.
• To assess the accuracy of this method on typical two-dimensional (2D) crystal structures.
• To explore the potential of quantum computing in advancing materials science simulations.

Main Methods:

• Implementation of a hybrid quantum machine learning algorithm utilizing restricted Boltzmann machines and a quantum algorithm.
• Benchmark testing conducted on the IBM-Q quantum computer.
• Calculation of electronic structures, specifically band structures, for hexagonal boron nitride and graphene.

Main Results:

• The hybrid quantum machine learning approach yielded band structures in strong agreement with conventional computational methods.
• Successful benchmark test demonstrates the feasibility of the implemented QML strategy.
• Validation of the accuracy for calculating electronic structures of 2D periodic systems.

Conclusions:

• The hybrid quantum machine learning method shows significant promise for accurate electronic structure calculations.
• Quantum-empowered computational approaches could revolutionize the study of quantum many-body systems.
• This work validates QML as a viable tool for materials science and condensed matter physics.