Parametrized quantum circuit for weight-adjustable quantum loop gas Rong-Yang Sun1 2Tomonori Shirakawa1 2and Seiji Yunoki1 2 3 4 1Computational Materials Science Research Team

2025-05-02 0 0 987.65KB 6 页 10玖币
侵权投诉
Parametrized quantum circuit for weight-adjustable quantum loop gas
Rong-Yang Sun,1, 2 Tomonori Shirakawa,1, 2 and Seiji Yunoki1, 2, 3, 4
1Computational Materials Science Research Team,
RIKEN Center for Computational Science (R-CCS), Kobe, Hyogo, 650-0047, Japan
2Quantum Computational Science Research Team,
RIKEN Center for Quantum Computing (RQC), Wako, Saitama, 351-0198, Japan
3Computational Quantum Matter Research Team,
RIKEN Center for Emergent Matter Science (CEMS), Wako, Saitama 351-0198, Japan
4Computational Condensed Matter Physics Laboratory,
RIKEN Cluster for Pioneering Research (CPR), Saitama 351-0198, Japan
(Dated: January 30, 2023)
Motivated by the recent success of realizing the topologically ordered ground state of the exactly
solvable toric code model by a quantum circuit on the real quantum device [K. J. Satzinger et al.,
Science 374, 1237 (2021)], here we propose a parametrized quantum circuit (PQC) with the same
real-device-performable optimal structure to represent quantum loop gas states with adjustably
weighted loop configurations. Combining such a PQC with the variational quantum eigensolver,
we obtain the accurate quantum circuit representation for the toric code model in an external
magnetic field with any field strength, where the system is not exactly solvable. The topological
quantum phase transition in this system is further observed in the optimized circuits by measuring
the magnetization and topological entanglement entropy.
Introduction.— The topologically ordered state, a new
category of exotic quantum states of matter, has continu-
ously attracted research interest in recent decades [1–3].
Many novel properties that emerge from it include the
degenerate ground state with long-range quantum en-
tanglement and the anyonic-type excitations, which are
also deemed essential for the development of quantum
computing [4, 5]. However, because of the existence of
the long-range quantum entanglement, it is challenging
to investigate the topologically ordered state. Except
for rare elaborately constructed exactly solvable models,
such as Kitaev’s toric code model [4] and Wen’s plaque-
tte model [6], it is difficult to verify whether a topologi-
cally ordered phase exists in a certain microscopic quan-
tum many-body Hamiltonian and to identify its nature,
a famous example being the spin-1/2 antiferromagnetic
Heisenberg model on the kagome lattice [7–10].
On the other hand, rapidly developed synthetic con-
trollable quantum platforms, both in circuit-based quan-
tum computers [11, 12] and analog quantum simula-
tors [13–15], offer a new way to explore these long-range-
entangled states. These controllable quantum systems
have an intrinsic quantum entangled nature and thus are
expected to be able to handle another quantum entangled
system relatively easier. Recently, this expectation has
been partially realized. One can use the optimal quan-
tum resource to realize the ground state of the toric code
model and accurately probe its properties [16], demon-
strating a huge potential for utilizing quantum computers
to study topologically ordered states.
Many algorithms have been proposed to efficiently use
the quantum resource for solving a generic non-exactly
solvable quantum many-body problem. Among them,
the most promising one, which is considered to be realiz-
able on current noisy-intermediate scale quantum (NISQ)
devices [17, 18], is the variational quantum eigensolver
(VQE) [19–21]. The VQE adopts a parametrized quan-
tum circuit (PQC) as the ansatz state. This PQC ansatz
is evaluated on a quantum computer, while its parame-
ters are optimized by a classical optimizer. Despite con-
siderable efforts, the scalable realization of the VQE on
NISQ devices is still under extensive research. This is be-
cause proper PQC ansatze, which can faithfully describe
the ground state of a given quantum many-body Hamil-
tonian and yet can be performed accurately enough on
real devices, are still missing, especially when the sys-
tem is close to a phase transition point (for example, see
Ref. [22]).
In this paper, we address this issue by proposing
a PQC, which has adequate expressibility towards the
ground state of a non-exactly solvable correlated Hamil-
tonian and at the same time can be faithfully evalu-
ated on NISQ devices. We encode a quantum loop gas
state with adjustably weighted loop configurations into a
quantum circuit with the optimal circuit depth. The ad-
justable weights are controlled by circuit parameters. Us-
ing such a single PQC, we precisely reproduce the ground
state with an energy accuracy better than the order of
102for the toric code model in an external magnetic
field, thus in both topologically ordered and ferromag-
netically ordered phases, by VQE calculations.
Toric code model in a magnetic field.— The toric code
model [4] is defined by qubits located on the bond centers
of a Lx×Lysquare lattice with Nbonds, and is given
by the following Hamiltonian:
HTC =X
s
AsX
p
Bp,(1)
where As=Qisσz
iand Bp=Qipσx
i. Here, σz
i(σx
i) is
the Pauli Z(X) operator of qubit ilocated on the i-th
arXiv:2210.14662v2 [quant-ph] 27 Jan 2023
2
(a)
(b)
Bp
AsB
Ψ(θ)=|+++···
||
|
FIG. 1. (a) Schematic figure of the toric code model in
an external magnetic field Bapplied along the zdirection
on the 6 ×5 lattice under OBCs, where qubits (indicated by
circles) are located on each bond of the lattice. Here, the
blue cross connecting four blue circles indicates an Asterm
and the green square connecting four green circles denotes a
Bpterm. Note that Asterms are also present at the corners
and edges of the lattice under OBCs, thus involving only two
and three qubits, respectively. The loops (indicated by red)
are formed by local qubit states |1ii, while qubits not along
any loops are |0ii. (b) Schematic figure of the ansatz state
|Ψ(θ)igenerated by a parametrized loop gas circuit. In these
two figures, the shade intensity of red for loops represents a
weight of the corresponding loops.
bond center, and s(p) sums over all the vertices (pla-
quettes) of the underlying square lattice [see Fig. 1(a)].
The Hamiltonian in Eq. (1) is exactly solvable, and its
ground state is a topological quantum spin liquid char-
acterized by a Z2topological order [4, 23]. For the toric
code model with open boundary conditions (OBCs) along
the two prime directions, its unique ground state can be
constructed as [16]
|Ψ0i=
Np
Y
p=1 1
2
Ip+1
2Bp|00 ···0i,(2)
where Ip=QipIiis the identity operator on the p-th
plaquette (Iibeing the identity operator at qubit i), Np
is the total number of the plaquettes, and |00 ···0i=
|0i1|0i2···|0iNrepresents a product of the local states
|0iiwith σz
i|0ii=|0iiand σz
i|1ii=−|1ii. Equation (2)
can be interpreted as generating an equally weighted su-
perposition of all possible basis configurations which con-
tain only closed loops formed by local states |1ii, i.e., a
quantum loop gas (LG).
An external magnetic field along the zdirection can
drive the toric code model away from the exactly solvable
point and leads to a topological quantum phase transi-
tion [24, 25]. The corresponding Hamiltonian [also see
Fig. 1(a)] is described by
HTCM(x) = (1 x)HTC x
N
X
i=1
σz
i.(3)
When x= 0, HTCM(x) returns to the toric code case,
while it has the exact ground state |00 ···0iwhen x= 1.
Extensive quantum Mount Carlo studies have suggested
that HTCM(x) goes through a second-order quantum
phase transition at xc0.25 from the toric code state
to the ferromagnetically ordered state in the thermody-
namic limit [25].
It is quite insightful to understand the ground state of
HTCM(x) using the LG picture and consider the mag-
netic field inducing tension to the loops [24]. In the
presence of the magnetic field, the loop with a more ex-
tended perimeter costs more energy, and its weight in
the ground state should be lightened to minimize the en-
ergy, as schematically illustrated in Fig. 1(a). Therefore,
an LG with adjustable weights for each loop pattern can
be regarded as an appropriate description of the ground
state of HTCM(x), especially when the charge excitation
is gapped out and thus closed loop configurations remain
without having open strings. In the rest of this paper,
we will show how to represent this as a PQC with an
optimal circuit structure.
Parametrized loop gas circuit.— Before introducing the
PQC for a weight-adjustable LG, we briefly recall the
method to construct the ground state |Ψ0iof the pure
toric code model HTC with an optimal quantum cir-
cuit [16, 26]. The core steps to generate |Ψ0iare to
apply a serial of (Ip+Bp)/2 operators, which can be
realized by applying a Hadamard gate Hto a representa-
tive qubit associated with a plaquette and then applying
CNOT gates controlled by the representative qubit to the
other qubits in this plaquette. The circuit depth in this
construction grows linearly with Ly, meeting the lower
bound of the information spread in the quantum cir-
cuit to construct a globally entangled quantum state [27].
Furthermore, the circuit constructed by this method has
been realized on the current NISQ devices with high ac-
curacy [16].
Applying a Hadamard gate to |0iiis nothing but creat-
ing the equally weighted superposition between |0ii(i.e.,
vacuum) and |1ii(a precursor of the loop). In order to
construct an LG with differently weighted loop patterns,
we need to create an imbalanced superposition between
|0iiand |1ii, which is rather straightforward once we no-
tice the relation
H|0ii=Ry(π/2)|0ii,(4)
where
Ry(θ) = cos θ
2sin θ
2
sin θ
2cos θ
2(5)
摘要:

Parametrizedquantumcircuitforweight-adjustablequantumloopgasRong-YangSun,1,2TomonoriShirakawa,1,2andSeijiYunoki1,2,3,41ComputationalMaterialsScienceResearchTeam,RIKENCenterforComputationalScience(R-CCS),Kobe,Hyogo,650-0047,Japan2QuantumComputationalScienceResearchTeam,RIKENCenterforQuantumComputing(...

展开>> 收起<<
Parametrized quantum circuit for weight-adjustable quantum loop gas Rong-Yang Sun1 2Tomonori Shirakawa1 2and Seiji Yunoki1 2 3 4 1Computational Materials Science Research Team.pdf

共6页,预览2页

还剩页未读, 继续阅读

声明:本站为文档C2C交易模式,即用户上传的文档直接被用户下载,本站只是中间服务平台,本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私,请立即通知玖贝云文库,我们立即给予删除!
分类:图书资源 价格:10玖币 属性:6 页 大小:987.65KB 格式:PDF 时间:2025-05-02

开通VIP享超值会员特权

  • 多端同步记录
  • 高速下载文档
  • 免费文档工具
  • 分享文档赚钱
  • 每日登录抽奖
  • 优质衍生服务
/ 6
客服
关注