Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 5
Encontrada(s) 1 página(s)

DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

The role of data embedding in equivariant quantum convolutional neural networks

  • Das, Sreetama
  • Martina, Stefano
  • Caruso, Filippo
Geometric deep learning refers to the scenario in which the symmetries of a dataset are used to constrain the parameter space of a neural network and thus, improve their trainability and generalization. Recently, this idea has been incorporated into the field of quantum machine learning, which has given rise to equivariant quantum neural networks (EQNNs). In this work, we investigate the role of classical-to-quantum embedding on the performance of equivariant quantum convolutional neural networks (EQCNNs) for the classification of images. We discuss the connection between the data embedding method and the resulting representation of a symmetry group and analyze how changing representation affects the expressibility of an EQCNN. We numerically compare the classification accuracy of EQCNNs with three different basis-permuted amplitude embeddings to the one obtained from a non-equivariant quantum convolutional neural network (QCNN). Our results show a clear dependence of classification accuracy on the underlying embedding, especially for initial training iterations. The improvement in classification accuracy of EQCNN over non-equivariant QCNN may be present or absent depending on the particular embedding and dataset used. The noisy simulation using simple noise models shows that certain EQCNNs are more robust to noise than non-equivariant QCNNs. It is expected that the results of this work can be useful to the community for a better understanding of the importance of data embedding choice in the context of geometric quantum machine learning., This work was financially supported by the European Union’s Horizon 2020 research and innovation programme under FET-OPEN GA No. 828946-PATHOS and by the European Commission’s Horizon Europe Framework Programme under the Research and Innovation Action GA No.101070546-MUQUABIS. S.M. also acknowledges financial support from the PNRR MUR project PE0000023-NQSTI. F.C. also acknowledges financial support by the European Defence Agency under the project Q-LAMPS Contract No. B PRJ-RT-989 and by the MUR Progetti di Ricerca di Rilevante Interesse Nazionale (PRIN) Bando 2022- project n. 20227HSE83 (ThAI-MIA) funded by the European Union-Next Generation EU., Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Otros (Other)

The role of data embedding in equivariant quantum convolutional neural networks

  • Das, Sreetama
  • Martina, Stefano
  • Caruso, Filippo
Geometric deep learning refers to the scenario in which the symmetries of a dataset are used to constrain the parameter space of a neural network and thus, improve their trainability and generalization. Recently this idea has been incorporated into the field of quantum machine learning, which has given rise to equivariant quantum neural networks (EQNNs). In this work, we investigate the role of classical-to-quantum embedding on the performance of equivariant quantum convolutional neural networks (EQCNNs) for the classification of images. We discuss the connection between the data embedding method and the resulting representation of a symmetry group and analyze how changing representation affects the expressibility of an EQCNN. We numerically compare the classification accuracy of EQCNNs with three different basis-permuted amplitude embeddings to the one obtained from a non-equivariant quantum convolutional neural network (QCNN). Our results show a clear dependence of classification accuracy on the underlying embedding, especially for initial training iterations. The improvement in classification accuracy of EQCNN over non-equivariant QCNN may be present or absent depending on the particular embedding and dataset used. It is expected that the results of this work can be useful to the community for a better understanding of the importance of data embedding choice in the context of geometric quantum machine learning., This work was supported by the European Commission’s Horizon Europe Framework Programme under the Research and Innovation Action GA n. 101070546–MUQUABIS, by the European Union’s Horizon 2020 research and innovation programme under FET-OPEN GA n. 828946–PATHOS, by the European Defence Agency under the project Q-LAMPS Contract No B PRJ- RT-989, and by the MUR Progetti di Ricerca di Rilevante Interesse Nazionale (PRIN) Bando 2022 - project n. 20227HSE83 – ThAI-MIA funded by the Euro- pean Union - Next Generation EU. S.M. acknowledges financial support from PNRR MUR project PE0000023-NQSTI., No
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

An Efficient Quantum Algorithm for the Traveling Salesman Problem

  • Sharma, Anant
  • Deshpande, Nupur
  • Ghosh, Sanchita
  • Das, Sreetama
  • Roy, Shibdas
The traveling salesman problem is the problem of finding out the shortest route in a network of cities, that a salesman needs to travel to cover all the cities, without visiting the same city more than once. This problem is known to be N P -hard with a brute-force complexity of O(N N) or O(N 2N) for N number of cities. This problem is equivalent to finding out the shortest Hamiltonian cycle in a given graph, if at least one Hamiltonian cycle exists in it. Quantum algorithms for this problem typically provide with a quadratic speedup only, using Grover's search, thereby having a complexity of O(N N/2) or O(N N). We present a bounded-error quantum polynomial-time (BQP) algorithm for solving the problem, providing with an exponential speedup. The overall complexity of our algorithm is O(N 3 log(N)κ/ϵ + 1/ϵ3), where the errors ϵ are O(1/poly(N)), and κ is the not-too-large condition number of the matrix encoding all Hamiltonian cycles., S. D. acknowledges financial support from the European Union’s Horizon 2020 research and innovation program under FET-OPEN Grant Agreement No. 828946 (PATHOS) and from the European Commission’s Horizon Europe Framework Programme under Research and Innovation Action Grant Agreement No. 101070546 (MUQUABIS)., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Roadmap for Photonics with 2D Materials

  • García de Abajo, Francisco Javier
  • Basov, D. N.
  • Koppens, Frank H. L.
  • Polini, Marco
  • Gonçalves, P. A. D.
  • Jacob, Zubin
  • Schuck, P. J.
  • Mugarza, Aitor
  • Tizei, Luiz H. G.
  • Menon, Vinod
  • Deilmann, Thorsten
  • Abate, Yohannes
  • Chervy, Thibault
  • Dal Conte, Stefano
  • Lienau, Christoph
  • Sun, Zhipei
  • Rozema, Lee A.
  • Alù, Andrea
  • Cox, Joel D.
  • Soavi, Giancarlo
  • Wu, Shiwei
  • Song, Justin C. W.
  • Kaminer, Ido
  • Giessen, Harald
  • Stauber, Tobias
  • Alonso-González, Pablo
  • Groep, Jorik van de
  • Boltasseva, Alexandra
  • Shalaev, Vladimir M.
  • Petek, Hrvoje
  • Mkhitaryan, Vahagn
  • Dai, Qing
  • Sternbach, Aaron
  • Koperski, Maciej
  • Novotny, Lukas
  • Vitiello, Miriam Serena
  • Patanè, Amalia
  • Xia, Fengnian
  • Burch, Kenneth S.
  • Merino, Pablo
  • Cuxart, M. G.
  • Chávez-Angel, Emigdio
  • Krishna Kumar, Roshan
  • Sánchez Sánchez, Miguel
  • Álvarez-Pérez, Gonzalo
  • Martín-Moreno, Luis
  • Nikitin, Alexey Y.
  • Ortega, J. Enrique
et al., Triggered by advances in atomic-layer exfoliation and growth techniques, along with the identification of a wide range of extraordinary physical properties in self-standing films consisting of one or a few atomic layers, two-dimensional (2D) materials such as graphene, transition metal dichalcogenides (TMDs), and other van der Waals (vdW) crystals now constitute a broad research field expanding in multiple directions through the combination of layer stacking and twisting, nanofabrication, surface-science methods, and integration into nanostructured environments. Photonics encompasses a multidisciplinary subset of those directions, where 2D materials contribute remarkable nonlinearities, long-lived and ultraconfined polaritons, strong excitons, topological and chiral effects, susceptibility to external stimuli, accessibility, robustness, and a completely new range of photonic materials based on layer stacking, gating, and the formation of moiré patterns. These properties are being leveraged to develop applications in electro-optical modulation, light emission and detection, imaging and metasurfaces, integrated optics, sensing, and quantum physics across a broad spectral range extending from the far-infrared to the ultraviolet, as well as enabling hybridization with spin and momentum textures of electronic band structures and magnetic degrees of freedom. The rapid expansion of photonics with 2D materials as a dynamic research arena is yielding breakthroughs, which this Roadmap summarizes while identifying challenges and opportunities for future goals and how to meet them through a wide collection of topical sections prepared by leading practitioners., This work has been supported in part by the funding sources listed here (grant numbers in parentheses). Section 2: European Research Council (ERC) (101141220 QUEFES); Spanish Ministry of Science, Innovation, and Universities (MICIN) (CEX2019-000910-S). Section 3: Gordon and Betty Moore Foundation (GBMF) (12212); Generalitat de Catalunya and European Social Fund-FEDER. Section 4: Italian Ministry of University and Research (PRIN 2020 2020JLZ52N); European Union (EU) (101131579, 873028). Section 5: Portuguese Foundation for Science and Technology (UIDB/04650/2020, COMPETE 2020, PORTUGAL 2020, FEDER, PTDC/FIS-MAC/2045/2021); Independent Research Fund Denmark (IRFD) (2032-00045B); Danish National Research Foundation (DNRF) (165). Section 6: Office of Naval Research (ONR) (N00014231270). Section 7: U.S. Department of Energy, Office of Science, Basic Energy Sciences (DOE) (DE-SC0019443). Section 8: Catalan CERCA; MICIN (CEX2021-001214-S, PID2022-140845OB-C63, CNS2022-135658, TED2021-132388B-C41, PID2021-125309OA-100, TED2021-129416A-I00); EU (PRTR-C17.I1); Comunidad de Madrid (TEC-2024/TEC-459, SINMOLMAT-CM); CSIC (20226AT011); CSIC and Czech Academy of Sciences (CAS) (BILAT23033/CSIC-24-10); CAS (GAČR 22-18718S). Section 9: French National Agency for Research (ANR) (ANR-20-CE42-0020). Section 10: F.D.: Emmy Noether Program (ENP) (534078167); H.D.: GBMF (10694), ONR (N00014-21-1-2770); C.S.: German Research Society (DFG) (SPP 2244 project SCHN1376 14.2); V.M.M.: GBMF (12764), ONR (W911NF-22-1-0091). Section 11: T.D.: DFG (426726249, DE 2749/2-1, DE 2749/2-2); A.C.: DFG (314695032 (SFB 1277 B05), 417590517 (SFB 1415 B11), 390858490 (EXC 2147 ct.qmat)), ERC (101001764 CoulENGINE); K.S.T.: Novo Nordisk Foundation Challenge Programme 2021 (BIOMAG NNF21OC0066526); K.S.T.: VILLUM FONDEN (37789). Section 12: Y.A.: Air Force Office of Scientific Research (AFOSR) (FA9550-23-1-0375), GBMF (12246); V.K.S. and M.C.H.: Northwestern University (NSF DMR-2308691); M.T.: AFOSR (FA9550-23-1-0447), National Science Foundation (NSF) (DMREF 2323296). Section 13: X.C.: DOE (BES MSE QIS FWP #100740); T.F.H.: DOE (BES CSGB AMOS FWP #SCW0063); L.Y.: DOE (Q-NEXT DE-AC02-76SF00515). Section 14: EU (IR0000016, ID D2B8D520, CUP B53C22001750006, 101130384, HORIZONEIC-2023-PATHFINDEROPEN-01, QUONDENSATE, M4C2-2022LA3TJ8 - CUP D53D23002280006, MSCA-2021 101066108, NQSTI PE00000023-q-ANTHEM-CUP H43C22000870001). Section 15: DFG (Li 580/16-1, DE 3578/3-1, SFB1372/2-Sig01, RTG1885, INST 184/163-1, INST 184/164-1); Niedersächsisches Ministerium für Wissenschaft and Kultur and Volkswagen Foundation (SMART, DyNano, Wissenschaftsraum ElLiKo). Section 16: Research Council of Finland (RSF) (353364, 360411, 359009, 365686, 367808, 320167, PREIN); EU (MSCA-RISE-872049 IPN-Bio, 101155089, ALTOS); ERC (834742). Section 17: DOE (DE-SC0019443). Section 18: IRFD (0165-00051B); DNRF (165); ERC (101141220-QUEFES). Section 19: DFG (CRC 1375 NOA, 398816777 C4, IRTG 2675 ‘Meta-Active’ 437527638 A4); Federal Ministry for Education and Research (BMBF) (16KIS1792 SiNNER). J.W.: DFG (ENP 503985532, CRC 1277-314695032 A03, RTG 2905-502572516). Section 20: National Key Research and Development Program of China (NKRDPC) (2024YFA1409800); National Natural Science Foundation of China (NNSFC) (12034003). Section 21: Singapore Ministry of Education (MOE) (MOE- T2EP50222-0011, MOE- MOET32023-0003). Section 22: ERC (101125662 QinPINEM), Israel Science Foundation (385/23); Hellen Diller Quantum Center; I.K.: GBMF (11473 iQCE); S.T.: Israeli Academy of Science and Humanities (Yad Hanadiv Foundation Rothschild), Israel Council for Higher Education (VATAT-Quantum), Technion (Viterbi). Section 23: DFG; BMBF; Vector-Stiftung; Zeiss-Stiftung; EU. Section 24: MICIN (PID2023-146461NB-I00, CEX2023-001316-M). Section 25: J.D.: Beijing Natural Science Foundation (Z240005), NNSFC (12474027); A.A.: ONR, Simons Foundation (SF); J.DC. ONR MURI (N0001-23-1-2567); A.Y.N.: MICIN (PID2023-147676NB-I00), Basque Department of Education (PIBA-2023-1-0007); P.A.-G.: ERC (101044461 TWISTOPTICS), MICIN (PID2022-141304NB-I00). Section 26: A.Y.N., P.A.G., and L.M.-M.: MICIN (PID2023-147676NB-I00, PID2022-141304NB-I00, PID2023-148359NB-C); L.M.-M.: Aragon Government (Q-MAD);. P.A.-G.: ERC (101044461, TWISTOPTICS); A.A.: ONR (N00014-2212448);. J.D.C and T.L.: ONR MURI (N0001-23-1-2567). Section 27: Department of Defense; NSF; SF. Section 28: J.v.d.G.: Dutch Science Foundation (VI.Vidi.203.027), ERC (101116984); T.S.: 2D-TECH VINNOVA Competence Center (ref.2019-00068); D.J.: ONR (N00014-23-1-203). Section 29: I.V.B.: U.S. Army Research Office (W911NF2310206); A.B. and V.M.S.: DOE (DE-SC0017717); AFOSR (FA9550-20-1-0124). Section 30: Purdue team: ONR (N00014-21-1-2026); J.S.: NSF (Graduate Research Fellowship DGE-1842166). Section 31: H.P.: NSF (CHE-2102601); S.T.: CAS (YSBR-054). Section 32: ERC (101141220 QUEFES); MCINN (CEX2019-000910-S, PID2022-137569NB-C42). Section 33: RCF (353364, 360411, 359009, 365686, 367808, Flagship Programme 320167 PREIN); EU (MSCA-RISE-872049 IPN-Bio); Jane and Aatos Erkko Foundation; Technology Industries of Finland Centennial Foundation (Future Makers 2022); ERC (834742); NNSFC (12374167, 12422406); NKRDPC (2022YFA1403500). Section 34: NNSFC (51925203, U2032206, 52072083, 51972072, 2021YFA1201500). Section 35: Y.Z.: DOE (DE-SC-0022885). Section 36: MOE (EDUN C-33-18-279-V12, I-FIM, MOE-T2EP50122-0012); AFOSR and ONR (FA8655-21-1-7026); MAT-GDT Program at A*STAR (M24N4b0034). Section 37: Swiss National Science Foundation (200021-227862). Section 38: ERC (101157731 Terascan); EU (EXTREME IR 944735, Graphene Flagship core 3, 101070546, PE0000023-NQSTI. Section 39: Defense Science and Technology Laboratory and Engineering and Physical Sciences Research Council (EP/T019018/1, EP/X524967/1). Section 40: ONR; NSF; Government of Israel. Section 41: P.W.: AFOSR (FA8655-20-1-7030 PhoQuGraph, FA8655-23-1-7063 TIQI); Austrian Science Fund (FWF 10.55776/F71); Austrian Federal Ministry of Labour and Economy; Christian Doppler Research Association; L.A.R.: Erwin Schrödinger Center for Quantum Science & Technology (ESQ Discovery). Section 42: K.S.B.: NSF (EPMD EPMD-2211334), DOE (DE-SC0018675); L.Z.: DOE (DE-SC0024145), NSF CAREER (DMR-174774); V.M.M.: DOE (DE-SC0025302); X.X.: DOE (DE-SC0018171), AFOSR MURI (FA9550-19-1-0390)., With funding from the Spanish government through the "Severo Ochoa Centre of Excellence" accreditation (CEX2019-000910-S)., With funding from the Spanish government through the "Severo Ochoa Centre of Excellence" accreditation (CEX2021-001214-S)., With funding from the Spanish government through the "Unit of Excellence Maria de Maeztu" accreditation (CEX2023-001316-M)., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Investigating layer-selective transfer learning of quantum approximate optimization algorithm parameters for the Max-Cut problem

  • Venturelli, Francesco Aldo
  • Das, Sreetama
  • Caruso, Filippo
The quantum approximate optimization algorithm (QAOA) is a variational quantum algorithm (VQA) ideal for noisy intermediate-scale quantum (NISQ) processors, and is highly successful in solving combinatorial optimization problems (COPs). It has been observed that the optimal parameters obtained from one instance of a COP can be transferred to another instance, resulting in generally good solutions for the latter. In this work, we propose a refinement scheme in which only a subset of QAOA layers is optimized following parameter transfer, with a focus on the Max-Cut problem. Our motivation is to reduce the complexity of the loss landscape when optimizing all the layers of high-depth QAOA circuits, as well as to reduce the optimization time. We investigate the potential hierarchical roles of different layers and analyze how the approximation ratio scales with increasing problem size. Our findings indicate that the selective layer optimization scheme offers a favorable trade-off between solution quality and computational time, and can be more beneficial than full optimization at a lower optimization time., This work was supported by the European Union's Research and Innovation Programme Horizon Europe G.A. No. 101070546 (MUQUABIS), by the European Defence Agency under the project Q-LAMPS Contract No. B PRJ- RT-989, by the PNRR MUR project PE0000023-NQSTI, and by the MUR Progetti di Ricerca di Rilevante Interesse Nazionale (PRIN) Bando 2022, Project No. 20227HSE83 ThAI-MIA funded by the European Union - Next Generation EU., Peer reviewed