Quantum adiabatic algorithm design using reinforcement learning

被引:28
作者
Lin, Jian [1 ,2 ]
Lai, Zhong Yuan [1 ,2 ]
Li, Xiaopeng [1 ,2 ,3 ,4 ]
机构
[1] Fudan Univ, Dept Phys, State Key Lab Surface Phys, Shanghai 200433, Peoples R China
[2] Fudan Univ, Inst Nanoelect & Quantum Comp, Shanghai 200433, Peoples R China
[3] Collaborat Innovat Ctr Adv Microstruct, Nanjing 210093, Peoples R China
[4] Shanghai Res Ctr Quantum Sci, Shanghai 201315, Peoples R China
基金
中国国家自然科学基金; 上海市自然科学基金;
关键词
SUPREMACY;
D O I
10.1103/PhysRevA.101.052327
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
Quantum algorithm design plays a crucial role in exploiting the computational advantage of quantum devices. Here we develop a deep-reinforcement-learning based approach for quantum adiabatic algorithm design. Our approach is generically applicable to a class of problems with solution hard-to-find but easy-to-verify, e.g., searching and NP-complete problems. We benchmark this approach in Grover-search and 3-SAT problems, and find that the adiabatic algorithm obtained by our RL approach leads to significant improvement in the resultant success probability. In application to Grover search, our RL design automatically produces an adiabatic quantum algorithm that has the quadratic speedup. We find for all our studied cases that quantitatively the RL-designed algorithm has a better performance compared to the analytically constructed nonlinear Hamiltonian path when the encoding Hamiltonian is solvable, and that this RL-design approach remains applicable even when the nonlinear Hamiltonian path is not analytically available. In 3-SAT we find RL design has fascinating transferability-the adiabatic algorithm obtained by training on a specific choice of clause number leads to better performance consistently over the linear algorithm on different clause numbers. These findings suggest the applicability of reinforcement learning for automated quantum adiabatic algorithm design. Further considering the established complexity equivalence of circuit and adiabatic quantum algorithms, we expect the RL-designed adiabatic algorithm to inspire novel circuit algorithms as well. Our approach is potentially applicable to different quantum hardware from trapped ions and optical lattices to superconducting-qubit devices.
引用
收藏
页数:11
相关论文
共 37 条
  • [1] Adiabatic Quantum Computation Is Equivalent to Standard Quantum Computation
    Aharonov, Dorit
    van Dam, Wim
    Kempe, Julia
    Landau, Zeph
    Lloyd, Seth
    Regev, Oded
    [J]. SIAM REVIEW, 2008, 50 (04) : 755 - 787
  • [2] Adiabatic quantum computation
    Albash, Tameem
    Lidar, Daniel A.
    [J]. REVIEWS OF MODERN PHYSICS, 2018, 90 (01)
  • [3] Quantum supremacy using a programmable superconducting processor
    Arute, Frank
    Arya, Kunal
    Babbush, Ryan
    Bacon, Dave
    Bardin, Joseph C.
    Barends, Rami
    Biswas, Rupak
    Boixo, Sergio
    Brandao, Fernando G. S. L.
    Buell, David A.
    Burkett, Brian
    Chen, Yu
    Chen, Zijun
    Chiaro, Ben
    Collins, Roberto
    Courtney, William
    Dunsworth, Andrew
    Farhi, Edward
    Foxen, Brooks
    Fowler, Austin
    Gidney, Craig
    Giustina, Marissa
    Graff, Rob
    Guerin, Keith
    Habegger, Steve
    Harrigan, Matthew P.
    Hartmann, Michael J.
    Ho, Alan
    Hoffmann, Markus
    Huang, Trent
    Humble, Travis S.
    Isakov, Sergei V.
    Jeffrey, Evan
    Jiang, Zhang
    Kafri, Dvir
    Kechedzhi, Kostyantyn
    Kelly, Julian
    Klimov, Paul V.
    Knysh, Sergey
    Korotkov, Alexander
    Kostritsa, Fedor
    Landhuis, David
    Lindmark, Mike
    Lucero, Erik
    Lyakh, Dmitry
    Mandra, Salvatore
    McClean, Jarrod R.
    McEwen, Matthew
    Megrant, Anthony
    Mi, Xiao
    [J]. NATURE, 2019, 574 (7779) : 505 - +
  • [5] Probing many-body dynamics on a 51-atom quantum simulator
    Bernien, Hannes
    Schwartz, Sylvain
    Keesling, Alexander
    Levine, Harry
    Omran, Ahmed
    Pichler, Hannes
    Choi, Soonwon
    Zibrov, Alexander S.
    Endres, Manuel
    Greiner, Markus
    Vuletic, Vladan
    Lukin, Mikhail D.
    [J]. NATURE, 2017, 551 (7682) : 579 - +
  • [6] Transitionless quantum driving
    Berry, M. V.
    [J]. JOURNAL OF PHYSICS A-MATHEMATICAL AND THEORETICAL, 2009, 42 (36)
  • [7] Quantum simulations come of age
    Bloch, Immanuel
    [J]. NATURE PHYSICS, 2018, 14 (12) : 1159 - 1161
  • [8] Photonic implementation of boson sampling: a review
    Brod, Daniel J.
    Galvao, Ernesto F.
    Crespi, Andrea
    Osellame, Roberto
    Spagnolo, Nicolo
    Sciarrino, Fabio
    [J]. ADVANCED PHOTONICS, 2019, 1 (03):
  • [9] Co-designing a scalable quantum computer with trapped atomic ions
    Brown, Kenneth R.
    Kim, Jungsang
    Monroe, Christopher
    [J]. NPJ QUANTUM INFORMATION, 2016, 2
  • [10] Reinforcement Learning in Different Phases of Quantum Control
    Bukov, Marin
    Day, Alexandre G. R.
    Sels, Dries
    Weinberg, Phillip
    Polkovnikov, Anatoli
    Mehta, Pankaj
    [J]. PHYSICAL REVIEW X, 2018, 8 (03):