Show simple item record

dc.contributor.authorSteck, James E.
dc.contributor.authorBehrman, Elizabeth C.
dc.contributor.authorThompson, Nathan
dc.date.accessioned2019-07-31T15:29:49Z
dc.date.available2019-07-31T15:29:49Z
dc.date.issued2019-01-06
dc.identifier.citationSteck, James E.; Behrman, Elizabeth C.; Thompson, Nathan. Machine learning applied to programming quantum computers. AIAA 2019-0956 Session: Quantum Control and Machine Learning (Invited) Published Online:6 Jan 2019en_US
dc.identifier.isbn978-162410578-4
dc.identifier.urihttps://doi.org/10.2514/6.2019-0956
dc.identifier.urihttp://hdl.handle.net/10057/16504
dc.descriptionClick on the DOI link to access the article (may not be free).en_US
dc.description.abstractWe apply machine learning to “program” quantum computers, both in simulation and in experimental hardware. A major difficulty in quantum computing is developing effective algorithms that can be programmed on a quantum device. Our approach is to apply machine learning to learn the quantum computer parameters that will yield the desired computation instead of choosing pre-made quantum gates to do the processing. As a proof of concept, we apply machine learning to a 16 qubit quantum gate computer developed by IBM and to a topological quantum computer by Microsoft. Preliminary results are shown of machine learning results both in software simulation and on the actual IBM quantum hardware. Microsoft results are shown only using their simulation as hardware is still being built. A second demonstration is to then port quantum machine learning to a large SQUID array of 2000 qubits originally designed to solve binary optimization problems via quantum annealing. To demonstrate quantum machine learning on this larger scale, it is programmed via machine learning to anneal to various entangled and partially entangled states; investigating a basic building block of general quantum computing. Simulation results are presented along with a method to demonstrate in hardware on a superconducting flux qubit quantum annealing machine housed at the Quantum Artificial Intelligence Laboratory (QuAIL) at NASA’s Advanced Supercomputing facility. Targeted entangled states are the relatively easy GHZ states, the EPR as well as the more difficult W and other states. Using machine learning instead of programming paves the way to greatly expanding the quantum computing capabilities of quantum computing hardware currently available.en_US
dc.language.isoen_USen_US
dc.publisherAmerican Institute of Aeronautics and Astronauticsen_US
dc.relation.ispartofseriesAIAA Scitech 2019 Forum;Session: Quantum Control and Machine Learning (Invited)
dc.subjectAviationen_US
dc.subjectComputer circuitsen_US
dc.subjectComputer programmingen_US
dc.subjectComputer softwareen_US
dc.subjectLogic gatesen_US
dc.subjectMachine learningen_US
dc.subjectNASAen_US
dc.subjectQuantum channelen_US
dc.subjectQuantum opticsen_US
dc.subjectQubitsen_US
dc.subjectSQUIDsen_US
dc.titleMachine learning applied to programming quantum computersen_US
dc.typeConference paperen_US
dc.rights.holder© 2019 American Institute of Aeronautics and Astronauticsen_US


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record