002883804 001__ 2883804
002883804 005__ 20241210040245.0
002883804 0248_ $$aoai:cds.cern.ch:2883804$$pcerncds:FULLTEXT$$pcerncds:CERN:FULLTEXT$$pcerncds:CERN
002883804 0247_ $$2DOI$$9arXiv$$a10.1109/QCE57702.2023.00033$$qpublication
002883804 037__ $$9arXiv$$aarXiv:2310.02323$$cquant-ph
002883804 035__ $$9arXiv$$aoai:arXiv.org:2310.02323
002883804 035__ $$9Inspire$$aoai:inspirehep.net:2706061$$d2024-12-09T08:22:07Z$$h2024-12-10T03:00:06Z$$mmarcxml$$ttrue$$uhttps://inspirehep.net/api/oai2d
002883804 035__ $$9Inspire$$a2706061
002883804 041__ $$aeng
002883804 100__ $$aChang, Su Yeon$$uCERN$$uEcole Polytechnique, Lausanne$$vIT Department, European Organization for Nuclear Research (CERN), Geneva, Switzerland$$vLaboratory of Theoretical Physics of Nanosystems (LTPN), Institute of Physics, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland
002883804 245__ $$9arXiv$$aApproximately Equivariant Quantum Neural Network for $p4m$ Group Symmetries in Images
002883804 269__ $$c2023-10-03
002883804 260__ $$c2023-09-17
002883804 300__ $$a7 p
002883804 520__ $$9IEEE$$aQuantum Neural Networks (QNNs) are suggested as one of the quantum algorithms which can be efficiently simulated with a low depth on near-term quantum hardware in the presence of noises. However, their performance highly relies on choosing the most suitable architecture of Variational Quantum Algorithms (VQAs), and the problem-agnostic models often suffer issues regarding trainability and generalization power. As a solution, the most recent works explore Geometric Quantum Machine Learning (GQML) using QNNs equivariant with respect to the underlying symmetry of the dataset. GQML adds an inductive bias to the model by incorporating the prior knowledge on the given dataset and leads to enhancing the optimization performance while constraining the search space. This work proposes equivariant Quantum Convolutional Neural Networks (EquivQCNNs) for image classification under planar p4m symmetry, including reflectional and 90° rotational symmetry. We present the results tested in different use cases, such as phase detection of the 2D Ising model and classification of the extended MNIST dataset, and compare them with those obtained with the non-equivariant model, proving that the equivariance fosters better generalization of the model.
002883804 520__ $$9arXiv$$aQuantum Neural Networks (QNNs) are suggested as one of the quantum algorithms which can be efficiently simulated with a low depth on near-term quantum hardware in the presence of noises. However, their performance highly relies on choosing the most suitable architecture of Variational Quantum Algorithms (VQAs), and the problem-agnostic models often suffer issues regarding trainability and generalization power. As a solution, the most recent works explore Geometric Quantum Machine Learning (GQML) using QNNs equivariant with respect to the underlying symmetry of the dataset. GQML adds an inductive bias to the model by incorporating the prior knowledge on the given dataset and leads to enhancing the optimization performance while constraining the search space. This work proposes equivariant Quantum Convolutional Neural Networks (EquivQCNNs) for image classification under planar $p4m$ symmetry, including reflectional and $90^\circ$ rotational symmetry. We present the results tested in different use cases, such as phase detection of the 2D Ising model and classification of the extended MNIST dataset, and compare them with those obtained with the non-equivariant model, proving that the equivariance fosters better generalization of the model.
002883804 540__ $$3preprint$$aCC BY 4.0$$uhttp://creativecommons.org/licenses/by/4.0/
002883804 540__ $$3publication$$aCC-BY-4.0$$bIEEE$$uhttps://creativecommons.org/licenses/by/4.0/
002883804 595_D $$aQIS$$d2023-10-12$$sabs
002883804 65017 $$2arXiv$$acs.LG
002883804 65017 $$2SzGeCERN$$aComputing and Computers
002883804 65017 $$2arXiv$$acs.AI
002883804 65017 $$2SzGeCERN$$aComputing and Computers
002883804 65017 $$2arXiv$$aquant-ph
002883804 65017 $$2SzGeCERN$$aGeneral Theoretical Physics
002883804 690C_ $$aCERN
002883804 690C_ $$aARTICLE
002883804 700__ $$aGrossi, Michele$$uCERN$$vIT Department, European Organization for Nuclear Research (CERN), Geneva, Switzerland
002883804 700__ $$aSaux, Bertrand Le$$uEuropean Space Agency$$vΦ-lab European Space Agency, Italy
002883804 700__ $$aVallecorsa, Sofia$$uCERN$$vIT Department, European Organization for Nuclear Research (CERN), Geneva, Switzerland
002883804 773__ $$wC23-09-17.3
002883804 8564_ $$82499488$$s27955$$uhttp://cds.cern.ch/record/2883804/files/MNIST.png$$y00001 : Extended MNIST
002883804 8564_ $$82499489$$s495247$$uhttp://cds.cern.ch/record/2883804/files/MNIST_test.png$$y00003 : Extended MNIST
002883804 8564_ $$82499490$$s22828$$uhttp://cds.cern.ch/record/2883804/files/Ising.png$$y00000 : Ising model
002883804 8564_ $$82499491$$s1475241$$uhttp://cds.cern.ch/record/2883804/files/2310.02323.pdf$$yFulltext
002883804 8564_ $$82499492$$s506351$$uhttp://cds.cern.ch/record/2883804/files/Ising_test.png$$y00002 : Ising
002883804 8564_ $$82694462$$s691359$$uhttp://cds.cern.ch/record/2883804/files/Publication.pdf$$yFulltext
002883804 960__ $$a13
002883804 962__ $$b2883771$$k229-235$$nbellevue20230917
002883804 980__ $$aConferencePaper
002883804 980__ $$aARTICLE