[Quthco] Fwd: [Quantum_spain_TalentQ] Seminario Quantum Spain

--------------------------------------------------------------------------------- Daniel Manzano Quantum Thermodynamics and Quantum Computation Group University of Granada Facultad de Ciencias, Av. Fuentenueva s/n Granada 18071, Spain Phone: +34 958241000 Ext: 20569 https://ic1.ugr.es/members/dmanzano/

Begin forwarded message:

From: TalentQ Subject: [Quantum_spain_TalentQ] Seminario Quantum Spain Date: 23. September 2024 at 09:49:50 CEST To: Reply-To: TalentQ

An initiative by: https://quantumspain-project.us21.list-manage.com/track/click?u=4de6018b41dfd67affc3f2acb&id=ce968c7651&e=03f2b554f6 QUANTUM SPAIN SEMINAR September 24th | 4pm (CEST) | Online

Quantum Convolutional Neural Networks are (Effectively) Classically Simulable Pablo Bermejo, PhD student at DIPC (Spain) and LANL (USA)

Quantum Convolutional Neural Networks (QCNNs) are widely regarded as a promising model for Quantum Machine Learning (QML). In this work we tie their heuristic success to two facts. First, that when randomly initialized, they can only operate on the information encoded in low-bodyness measurements of their input states. And second, that they are commonly benchmarked on «locally-easy» datasets whose states are precisely classifiable by the information encoded in these low-bodyness observables subspace. We further show that the QCNN’s action on this subspace can be efficiently classically simulated by a classical algorithm equipped with Pauli shadows on the dataset. Indeed, we present a shadow-based simulation of QCNNs on up-to 1024 qubits for phases of matter classification. Our results can then be understood as highlighting a deeper symptom of QML: Models could only be showing heuristic success because they are benchmarked on simple problems, for which their action can be classically simulated. This insight points to the fact that non-trivial datasets are a truly necessary ingredient for moving forward with QML. To finish, we discuss how our results can be extrapolated to classically simulate other architectures.

Join Meeting: https://events.teams.microsoft.com/event/06fd2ebc-0541-4ec9-966e-8cf22ae1a68... https://quantumspain-project.us21.list-manage.com/track/click?u=4de6018b41dfd67affc3f2acb&id=0e6b9a8a43&e=03f2b554f6 https://quantumspain-project.us21.list-manage.com/track/click?u=4de6018b41dfd67affc3f2acb&id=aaa97ee777&e=03f2b554f6 REGISTRARSE https://quantumspain-project.us21.list-manage.com/track/click?u=4de6018b41dfd67affc3f2acb&id=e5e8c3b80a&e=03f2b554f6 https://quantumspain-project.us21.list-manage.com/track/click?u=4de6018b41dfd67affc3f2acb&id=434c54081b&e=03f2b554f6 https://quantumspain-project.us21.list-manage.com/track/click?u=4de6018b41dfd67affc3f2acb&id=321d9aed6d&e=03f2b554f6 Copyright (C) 2024 USC. Todos los derechos reservados. Has recibido este correo electrónico porque lo has aceptado en nuestro sitio web.

Nuestra dirección de correo es: USC Rua Nova 6 Santiago De Compostela, Galicia 15705 Spain

¿Quieres cambiar la forma en que recibes estos correos electrónicos? Puedes actualizar tus preferencias https://quantumspain-project.us21.list-manage.com/profile?u=4de6018b41dfd67affc3f2acb&id=58c57f70fc&e=03f2b554f6&c=171303603f o cancelar la suscripción https://quantumspain-project.us21.list-manage.com/unsubscribe?u=4de6018b41dfd67affc3f2acb&id=58c57f70fc&t=b&e=03f2b554f6&c=171303603f https://quantumspain-project.us21.list-manage.com/track/click?u=4de6018b41dfd67affc3f2acb&id=ff6fda3be2&e=03f2b554f6_______________________________________________ Quantum_Spain_TalentQ mailing list -- quantum_spain_talentq@bsc.es mailto:quantum_spain_talentq@bsc.es To unsubscribe send an email to quantum_spain_talentq-leave@bsc.es mailto:quantum_spain_talentq-leave@bsc.es