WO2020077521A1 - 量子状态搜索方法及装置 - Google Patents
量子状态搜索方法及装置 Download PDFInfo
- Publication number
- WO2020077521A1 WO2020077521A1 PCT/CN2018/110389 CN2018110389W WO2020077521A1 WO 2020077521 A1 WO2020077521 A1 WO 2020077521A1 CN 2018110389 W CN2018110389 W CN 2018110389W WO 2020077521 A1 WO2020077521 A1 WO 2020077521A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- quantum
- state
- iterative
- input state
- black box
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N10/00—Quantum computing, i.e. information processing based on quantum-mechanical phenomena
- G06N10/60—Quantum algorithms, e.g. based on quantum optimisation, quantum Fourier or Hadamard transforms
Definitions
- the invention relates to the technical field of information security, in particular to a quantum state search method and device.
- Quantum computing is the product of the combination of quantum mechanics and computer science, and it is a new type of computing method. Among them, the quantum state search method has attracted the attention of researchers because it breaks the limitations of some classic algorithms and improves the speed of the algorithm.
- the main purpose of the embodiments of the present invention is to provide a quantum state search method and device, which can improve the calculation rate of the quantum state search method.
- a first aspect of an embodiment of the present invention provides a quantum state search method.
- the method includes: transforming a quantum initial state according to a Hadamard transform effect to obtain a quantum input state; determining iterations based on a quantum black box and the quantum input state Operator, iterative operation based on the iterative operator and the quantum input state to obtain a quantum intermediate state, and determine the quantum intermediate state as a new quantum input state; execute the quantum black box and the quantum
- the input state determines an iterative operator, based on the iterative operator and iterative operation on the quantum input state to obtain a quantum intermediate state, and the step of determining the quantum intermediate state as a new quantum input state until reaching a preset Termination condition, the obtained quantum intermediate state is determined as the labeled quantum state.
- a second aspect of an embodiment of the present invention provides a quantum state search device.
- the device includes: a transformation unit for transforming a quantum initial state according to the Hadamard transformation effect to obtain a quantum input state; An iterative operator is determined by the black box and the quantum input state, an iterative operation is performed on the iterative operator and the quantum input state to obtain a quantum intermediate state, and the quantum intermediate state is determined as a new quantum input state; output A unit for performing the iterative operator determination based on the quantum black box and the quantum input state, performing iterative operations on the iterative operator and the quantum input state to obtain a quantum intermediate state, and converting the quantum intermediate state
- the state is determined as the step of the new quantum input state until the preset termination condition is reached, then the obtained quantum intermediate state is determined as the labeled quantum state.
- 1 is a schematic diagram of an implementation process of the quantum state search method in the first embodiment provided by the present invention
- FIG. 2 is a circuit block diagram of the quantum state search method in the first embodiment provided by the present invention.
- FIG. 3 is a circuit diagram of the iterative operator in the first embodiment provided by the present invention.
- FIG. 4 is a schematic diagram of performing two iteration operations in the first embodiment provided by the present invention.
- FIG. 5 is a schematic diagram of the application of the quantum state search method in the first embodiment provided by the present invention.
- FIG. 6 is a schematic structural diagram of a quantum state search device in a second embodiment provided by the present invention.
- FIG. 1 is a schematic diagram of an implementation process of the quantum state search method in the first embodiment provided by the present invention. As shown in Figure 1, the method mainly includes the following steps:
- quantum computing is a product of the combination of quantum mechanics and computer science.
- the quantum state search method in this embodiment is applied to a disordered database, and the data in the disordered database is randomly arranged.
- the problem is now described as searching for labeled quantum states in a search space that includes N quantum states to be searched.
- the quantum initial state is transformed according to the Hadamard transformation effect to obtain the quantum input state:
- the quantum input state is a superposition state of equal amplitude value, and the quantum input state is further expressed as:
- the quantum black box is in the form of a unitary matrix.
- the quantum black box is determined according to the preset target state by the following formula:
- O is the quantum black box
- I is the identity matrix
- ⁇ > is the preset target state.
- the iteration operator used in the first iteration operation is:
- G 0 (2
- the quantum intermediate state obtained after the first iteration operation is completed that is, the quantum input state used in the second iteration operation is:
- n qubits are the quantum initial state, and the oracle work space is a quantum black box.
- the quantum black box is in the form of a unitary matrix, and the following formula is used to determine an iterative operator based on the quantum black box and the quantum input state, and perform iterative operations on the iterative operator and the quantum input state to obtain a quantum intermediate state , And the step of determining the quantum intermediate state as a new quantum input state until the preset termination condition is reached, then the obtained quantum intermediate state is determined as the labeled quantum state:
- ⁇ 1 > obtained after the first iteration operation process is completed is used as the quantum input state used in the second iteration operation.
- the iteration operator used in the second iteration operation is:
- G 1 (2
- the change operation operator in the second iteration operation is:
- the quantum intermediate state obtained after the second iteration operation that is, the quantum input state used in the third iteration operation is:
- the iteration operator used in the third iteration process is:
- the change operation operator in the third iteration operation is:
- the quantum intermediate state after the completion of the ith iteration operation that is, the quantum input state used in the ith + 1 iteration operation is defined as:
- the obtained quantum intermediate state is the marked state
- the iterative operator used during the kth iteration operation is:
- G k-1 (2
- the steps of the quantum state search method for the multi-variable rotation axis depict the geometric change graph of its iterative process. Therefore, the process and results of describing the entire quantum state search method in a geometric manner are as follows:
- N is the total number of all quantum states to be searched
- M is the number of labeled quantum states
- ⁇ > is the non-target state
- ⁇ > is the target state
- ⁇ 0 > can be regarded as a vector in the two-dimensional space formed by the vector
- the quantum black box O in the quantum state search method is used to mark the target state, so this operation is equivalent to performing a reflection transformation on the plane formed by
- ⁇ >) a
- the quantum input state is transformed according to the iterative operator G 0 used in the first iteration operation, and the resulting quantum intermediate state is:
- the quantum intermediate state obtained after the third iteration is:
- the quantum intermediate state obtained after implementing the fourth iteration calculation process is:
- the quantum intermediate state obtained after the kth fall operation is:
- -I, then the operator U i + (2
- -I during the i + 1 iteration is a unitary operator, which satisfies that there is no energy loss in the quantum system, and all transformations are reversible transformations or unitary transformations. condition.
- the performance analysis of the quantum search method in this embodiment is as follows:
- ⁇ 0 > before the first iteration operation will become closer to the marked quantum state
- the labeled quantum state can be determined, and the angle meets the condition:
- the minimum number of iteration termination times to obtain the labeled quantum state in this embodiment is:
- CI (x) represents the closest integer to the real number x.
- the minimum number of iteration termination times to obtain the labeled quantum state in this embodiment is:
- the quantum state search method in this embodiment obtains the labeled quantum state with fewer iteration termination times, so the performance of the quantum state search method in this embodiment is better.
- the Grover algorithm and the quantum state search method in this embodiment act on the quantum state
- the initial angle ⁇ reaches ⁇ / 6.
- the G operation in the figure represents the iteration operator of the Grover algorithm
- G i represents the iteration operator of the quantum search method in this embodiment.
- ⁇ > rotates to the position of ⁇ / 6
- the Grover algorithm only rotates 2 ⁇ after another rotation operation.
- the quantum search method in this embodiment is close to the marked quantum state after another rotation operation, and its rotation angle is much larger than that of the Grover algorithm. This further illustrates that the quantum search method in this embodiment has an algorithm rate superior to the Grover algorithm. .
- the iterative operator used after the subsequent iterative operation changes as the number of iterations increases, and the application of the multi-variable iterative operator accelerates the increase of the target state probability amplitude Accelerate the reduction of the non-target state probability amplitude.
- the rotation axis also changes as the number of iterations increases, the rotation angle is 3 k ⁇ , instead of using a fixed rotation axis, thus breaking through the limitation of the calculation rate of the original quantum search algorithm and improving the search The calculation rate of the labeled quantum state.
- FIG. 6 is a schematic structural diagram of a quantum state search device in a second embodiment provided by the present invention. As shown in Figure 6, the device mainly includes:
- the transformation unit 201 is configured to transform the quantum initial state according to the Hadamard transformation effect to obtain a quantum input state.
- the iteration unit 202 is used to determine an iteration operator according to the quantum black box and the quantum input state, perform an iterative operation on the iteration operator and the quantum input state to obtain a quantum intermediate state, and determine the quantum intermediate state as the new quantum input state.
- the output unit 203 is used to perform the iterative operator determination based on the quantum black box and the quantum input state, perform iterative operations on the iterative operator and the quantum input state to obtain the quantum intermediate state, and determine the quantum intermediate state as the new quantum input state Step until the preset termination condition is reached, the obtained quantum intermediate state is determined as the labeled quantum state.
- the quantum black box is in the form of a unitary matrix
- the output unit 203 is also used to execute an iterative operator based on the quantum black box and the quantum input state through the following formula, perform an iterative operation on the iterative operator and the quantum input state to obtain a quantum intermediate state, and determine the quantum intermediate state as The steps of the new quantum input state until the preset termination condition is reached, then the obtained quantum intermediate state is determined as the labeled quantum state:
- the transformation unit 201 is also used to transform the quantum initial state according to the Hadamard transformation effect according to the following formula to obtain a quantum input state:
- the device further includes: a determining unit 204.
- the determining unit 204 is configured to determine the quantum black box according to the preset target state.
- the determining unit 204 is also used to determine the quantum black box according to the preset target state by the following formula,
- O is the quantum black box
- I is the identity matrix
- ⁇ > is the preset target state.
- the iterative operator used after the subsequent iterative operation changes as the number of iterations increases, and the application of the multi-variable iterative operator accelerates the increase of the target state probability amplitude Accelerate the reduction of the non-target state probability amplitude.
- the rotation axis also changes as the number of iterations increases, and the rotation angle is 3 k ⁇ , instead of using a fixed rotation axis, thus breaking through the limitation of the calculation rate of the original quantum search algorithm and improving the search The calculation rate of the labeled quantum state.
Landscapes
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mathematical Analysis (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- Computational Mathematics (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Superconductor Devices And Manufacturing Methods Thereof (AREA)
Abstract
一种量子状态搜索方法及装置,应用于信息安全技术领域。该量子状态搜索方法包括:根据哈达玛变换效应将量子初始态进行变换,得到量子输入态(101)。根据量子黑盒和量子输入态确定迭代算子,基于迭代算子和量子输入态上进行迭代运算,得到量子中间态,并将量子中间态确定为新量子输入态(103)。执行根据量子黑盒和量子输入态确定迭代算子,基于迭代算子和量子输入态上进行迭代运算,得到量子中间态,并将量子中间态确定为新量子输入态的步骤,直至达到预设的终止条件,则得到的量子中间态确定为被标记的量子状态(104)。该方法可提高量子状态搜索方法的计算速率。
Description
本发明涉及信息安全技术领域,尤其涉及一种量子状态搜索方法及装置。
随着信息化社会的不断提高,微电子技术飞速发展,使得电路上元器件的尺寸越来越小。根据摩尔定律,传统计算机的计算能力会随着集成电路上元器件数目的增加而增加。然而进入纳米时代,晶体管的体积越来越小,电路上增加的元器件越来越多,会严重影响晶体管的性能。其次,小尺寸晶体管会带来量子隧穿效应,会干扰经典计算机的计算,若处理不好会烧坏电子线路。电路的高集成度导致摩尔定律渐渐无效,使传统计算机的发展遇到瓶颈。因此,量子计算凭借其强大的并行计算能力成为国际的研究热点。量子计算是量子力学与计算机科学结合的产物,是一种新型的计算方式。其中,量子状态搜索方法由于打破某些经典算法的限制,提升了算法的速率而受到了广大研究者的关注。
然而,目前的Grover量子搜索算法的计算速率存在不够高的问题。
发明内容
本发明实施例的主要目的在于提供一种量子状态搜索方法及装置,可提高量子状态搜索方法的计算速率。
本发明实施例第一方面提供了一种量子状态搜索方法,所述方法包括:根据哈达玛变换效应将量子初始态进行变换,得到量子输入态;根据量子黑盒和所述量子输入态确定迭代算子,基于所述迭代算子和所述量子输入态上进行迭 代运算,得到量子中间态,并将所述量子中间态确定为新量子输入态;执行所述根据量子黑盒和所述量子输入态确定迭代算子,基于所述迭代算子和所述量子输入态上进行迭代运算,得到量子中间态,并将所述量子中间态确定为新量子输入态的步骤,直至达到预设的终止条件,则得到的量子中间态确定为被标记的量子状态。
本发明实施例第二方面提供了一种量子状态搜索装置,所述装置包括:变换单元,用于根据哈达玛变换效应将量子初始态进行变换,得到量子输入态;迭代单元,用于根据量子黑盒和所述量子输入态确定迭代算子,基于所述迭代算子和所述量子输入态上进行迭代运算,得到量子中间态,并将所述量子中间态确定为新量子输入态;输出单元,用于执行所述根据量子黑盒和所述量子输入态确定迭代算子,基于所述迭代算子和所述量子输入态上进行迭代运算,得到量子中间态,并将所述量子中间态确定为新量子输入态的步骤,直至达到预设的终止条件,则得到的量子中间态确定为被标记的量子状态。
从上述实施例可知,通过利用多变的迭代算子,使得后续迭代运算后使用的迭代算子随着迭代次数的增加而变换,并且在搜索过程中,旋转轴同样随着迭代次数的增加而变换,而不是利用固定的迭代算子和固定的旋转轴,从而突破了原有量子搜索算法计算速率的限制,提高了搜索被标记的量子状态的计算速率。
图1是本发明提供的第一实施例中的量子状态搜索方法的实现流程示意图;
图2是本发明提供的第一实施例中的量子状态搜索方法的线路框图;
图3是本发明提供的第一实施例中的迭代算子的线路图;
图4是本发明提供的第一实施例中的进行两次迭代运算的示意图;
图5是本发明提供的第一实施例中的量子状态搜索方法的应用示意图;
图6是本发明提供的第二实施例中的量子状态搜索装置的结构示意图。
为使得本发明的发明目的、特征、优点能够更加的明显和易懂,下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而非全部实施例。基于本发明中的实施例,本领域技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
请参阅图1,图1是本发明提供的第一实施例中的量子状态搜索方法的实现流程示意图。如图1所示,该方法主要包括以下步骤:
101、根据哈达玛变换效应将量子初始态进行变换,得到量子输入态。
具体的,量子计算是量子力学与计算机科学结合的产物,本实施例中的量子状态搜索方法应用于无序数据库,无序数据库中的数据呈现随机排布。现将问题描述为在包括N个待搜索量子状态的搜索空间中搜索被标记的量子状态。
进一步地,通过以下公式,根据Hadamard(哈达玛)变换效应将量子初始态进行变换,得到量子输入态:
102、根据预设目标状态确定量子黑盒。
具体的,该量子输入态为等幅度值的叠加态,量子输入态进一步表示为:
该量子黑盒为酉矩阵形式,通过以下公式,根据预设目标状态确定量子黑盒:
O=I-2|τ><τ|。
式中,O为量子黑盒,I为单位矩阵,|τ>为预设目标状态。
103、根据量子黑盒和该量子输入态确定迭代算子,基于该迭代算子和该量子输入态上进行迭代运算,得到量子中间态,并将该量子中间态确定为新量子输入态。
具体的,第一次迭代运算中使用的迭代算子为:
G
0=(2|ψ
0><ψ
0|-I)O。
则第一次迭代运算完成后得到的量子中间态,即第二次迭代运算中使用的量子输入态为:
104、执行该根据量子黑盒和该量子输入态确定迭代算子,基于该迭代算子和该量子输入态上进行迭代运算,得到量子中间态,并将该量子中间态确定为新量子输入态的步骤,直至达到预设的终止条件,则得到的量子中间态确定为被标记的量子状态。
具体的,如图2及图3所示,n量子比特为量子初始态,oracle的工作空间为量子黑盒。
进一步地,量子黑盒为酉矩阵形式,通过以下公式,执行该根据量子黑盒和该量子输入态确定迭代算子,基于该迭代算子和该量子输入态上进行迭代运算,得到量子中间态,并将该量子中间态确定为新量子输入态的步骤,直至达到预设的终止条件,则得到的量子中间态确定为被标记的量子状态:
|ψ
k>=G
k-1G
k-2…G
1G
0|ψ
0>;
G
i=(2|ψ
i><ψ
i|-I)O;
式中,|ψ
k>为被标记的量子状态,|ψ
i>为第i+1次迭代运算中使用的量子输入态,G
i为i+1次迭代运算中使用的迭代算子,i∈[0,k-1],k为迭代终止次数,i∈[0,k-1],O为量子黑盒。
其中,将第一次迭代运算过程完成后得到的量子中间态|ψ
1>作为第二次迭代 运算中使用的量子输入态,第二次迭代运算中使用的迭代算子为:
G
1=(2|ψ
1><ψ
1|-I)O。
其中,第二次迭代运算中的变化操作算子为:
则第二次迭代运算完成后得到的量子中间态,即第三次迭代运算中使用的量子输入态为:
接着,第三次迭代过程中使用的迭代算子为:
G
2=(2|ψ
2><ψ
2|-I)O。
其中,第三次迭代运算中的变化操作算子为:
则第三次迭代运算完成后得到的量子中间态,第四次迭代运算中使用的量子输入态为:
从上述|ψ
1>、|ψ
2>及|ψ
3>的表达式中可以看到,在每次迭代运算过程中使用的迭代算子并不固定。就每次迭代之后的状态而言,多变迭代算子的运用加快了预设目标状态概率幅的增加,加快了非预设目标状态概率幅的降低。
因此,定义第i次迭代运算完成后的量子中间态,即第i+1次迭代运算中使用的量子输入态为:
|ψ
i>=a|ψ
0>+b|τ>(0<a<1,0<b<1)。
则在第i+1次迭代运算中使用的迭代算子为:
G
i=(2|ψ
i><ψ
i|-I)O。
其中,第i+1次迭代运算中的变化操作算子为:
2|ψ
i><ψ
i|-I=2(a|ψ
0>+b|τ>)(a<ψ
0|+b<τ|)-I
=2a
2|ψ
0><ψ
0|+2ab|ψ
0><τ|+2ab。
|τ><ψ
0|+2b
2|τ><τ|-I
则第i+1次迭代运算完成后得到的量子中间态为:
根据上述的迭代公式,经过k次迭代后,得到的量子中间态为被标记的状态,在第k次迭代运算过程中使用的迭代算子为:
G
k-1=(2|ψ
k-
1><ψ
k-
1|-I)O。
则在第k次迭代运算完成后得到的量子中间态,即被标记的量子状态为:
可选的,针对多变旋转轴的量子状态搜索方法的步骤描绘了其迭代过程的几何变化图,因此,通过几何方式描述整个量子状态搜索方法的过程及结果如下所述:
因此,|ψ
0>可看作向量|α>和向量|β>所形成的二维空间里的向量。而在量子状态搜索方法中的量子黑盒O用于标记目标状态,因此,该运算相当于为|α>和|β>所形成的平面上相对于向量|α>实施了一次反射变换,即O(a|α>+b|β>)=a|α>-b|β>。
其中,在第i+1次迭代运算过程中使用的迭代算子为:G
i=(2|ψ
i><ψ
i|-I)O, 因
此,相当于在|α>和|β>所形成的平面上,运算2|ψ
i><ψ
i|-I实施了一次相当于向量|ψ
i>的反射变换。然而,这两次的反射变换操作形成了一次旋转操作。
|ψ
0>=cosθ|α>+sinθ|β>。
如图4所示,根据第一次迭代运算中使用的迭代算子G
0将量子输入态进行变换,得到的量子中间态为:
|ψ
1>=G
0|ψ
0>=cos3θ|α>+sin3θ|β>。
由于在第二次迭代运算过程中使用的迭代算子不再是G
0,因此,不再以向量|ψ
0>为对称轴进行反射,而是以第一次迭代运算完成后得到的量子中间态|ψ
1>进行反射,因此,第二次迭代运算后得到的量子中间态为:
|ψ
2>=G
1|ψ
1>=G
1G
0|ψ
0>=cos9θ|α>+sin9θ|β>。
实施第三次迭代运算后得到的量子中间态为:
|ψ
3>=G
2|ψ
2>=G
2G
1G
0|ψ
0>=cos27θ|α>+sin27θ|β>。
实施第四次迭代运算过程后得到的量子中间态为:
|ψ
4>=G
3|ψ
3>=G
3G
2G
1G
0|ψ
0>=cos81θ|α>+sin81θ|β>。
依照上述规律,实施第k次跌倒操作后得到的量子中间态为:
|ψ
k>=G
k-1|ψ
k-1>=G
k-1G
k-2…G
1G
0|ψ
0>=cos(3
k)θ|α>+sin(3
k)θ|β>。
可选的,本实施例中的量子状态搜索方法的可行性证明如下:
而第i+1次迭代过程中使用的迭代算子为:G
i=(2|ψ
i><ψ
i|-I)O,其中O为酉算子。其中第i+1次迭代过程中的变化操作算子为:U
i=2|ψ
i><ψ
i|-I,则算子U
i
+=(2|ψ
i><ψ
i|-I)
+。
其中,U
iU
i
+=(2|ψ
i><ψ
i|-I)(2|ψ
i><ψ
i|-I)
+=4|ψ
i><ψ
i|-2|ψ
i><ψ
i|-2|ψ
i><ψ
i|+I=I。
因此,第i+1次迭代过程中的变化操作算子2|ψ
i><ψ
i|-I为酉算子,满足量子系统中不存在能量损耗,所有变换均为可逆变换即酉变换的条件。
可选的,本实施例中的量子搜索方法的性能分析如下:
通过比较使用固定的迭代算子的量子搜索算法Grover和本实施例中的量子搜索方法的变换规律,在经过k次迭代后,二者的旋转角度的变化情况,如表1所示:
表1
| 算法\迭代次数 | 0 | 1 | 2 | 3 | … | k-1 | k |
| Grover算法 | θ | 3θ | 5θ | 7θ | … | (2k-1)θ | (2k+1)θ |
| 多变旋转算法 | θ | 3θ | 9θ | 27θ | … | 3 k-1θ | 3 kθ |
根据上述表格可以看到,随着迭代次数的增加,第一次迭代运算前的量子输入态|ψ
0>会越来越接近被标记的量子状态|β>。当经过k次迭代后,则可确定被标记的量子状态,此时角度满足条件:|sin(3
k)θ|≈1。
因此,在本实施例中得到被标记的量子状态的最少的迭代终止次数为:
因此,在本实施例中得到被标记的量子状态的最少的迭代终止次数为:
而使用固定的迭代算子的量子搜索算法的最少的迭代终止次数为:
通过比较k和k
G,可以得到,本实施例中的量子状态搜索方法以更少的迭代终止次数得到被标记的量子状态,因此本实施例中的量子状态搜索方法的性能更好。
如图5所示,Grover算法与本实施例中的量子状态搜索方法同时作用于量子状态|ψ>,分别经过k'
1和k'
2次相应的旋转操作后,使初始角度θ达到π/6。图中G操作表示Grover算法的迭代算子,G
i表示本实施例中的量子搜索方法的迭代算子。初始状态|ψ>旋转到在π/6的位置时,Grover算法再经过一次旋转操作 只旋转了2θ。而本实施例中的量子搜索方法再经过一次旋转操作已经接近被标记的量子状态,其旋转角度远大于Grover算法的旋转角度,进一步说明本实施例中的量子搜索方法的算法速率优于Grover算法。
在本发明实施例中,通过利用多变迭代算子,使得后续迭代运算后使用的迭代算子随着迭代次数的增加而变换,多变迭代算子的运用加快了目标状态概率幅的增加,加快了非目标状态概率幅的降低。并且在几何图示中,旋转轴同样随着迭代次数的增加而变换,旋转角度为3
kθ,而不是利用固定的旋转轴,从而突破了原有量子搜索算法计算速率的限制,提高了搜索被标记的量子状态的计算速率。
参见图6,图6是本发明提供的第二实施例中的量子状态搜索装置的结构示意图。如图6所示,该装置主要包括:
变换单元201,用于根据哈达玛变换效应将量子初始态进行变换,得到量子输入态。
迭代单元202,用于根据量子黑盒和量子输入态确定迭代算子,基于迭代算子和量子输入态上进行迭代运算,得到量子中间态,并将量子中间态确定为新量子输入态。
输出单元203,用于执行根据量子黑盒和量子输入态确定迭代算子,基于迭代算子和量子输入态上进行迭代运算,得到量子中间态,并将量子中间态确定为新量子输入态的步骤,直至达到预设的终止条件,则得到的量子中间态确定为被标记的量子状态。
进一步地,所述量子黑盒为酉矩阵形式,则,
输出单元203,还用于通过以下公式,执行根据量子黑盒和量子输入态确定迭代算子,基于迭代算子和量子输入态上进行迭代运算,得到量子中间态,并将量子中间态确定为新量子输入态的步骤,直至达到预设的终止条件,则得到的量子中间态确定为被标记的量子状态:
|ψ
k>=G
k-1G
k-2…G
1G
0|ψ
0>。
G
i=(2|ψ
i><ψ
i|-I)O。
式中,|ψ
k>为被标记的量子状态,|ψ
i>为第i+1次迭代运算中使用的量子输入态,G
i为i+1次迭代运算中使用的迭代算子,i∈[0,k-1],k为迭代终止次数,i∈[0,k-1],O为量子黑盒。
进一步地,变换单元201,还用于通过以下公式,根据哈达玛变换效应将量子初始态进行变换,得到量子输入态:
进一步地,所述量子输入态为等幅度值的叠加态,则装置还包括:确定单元204。
确定单元204,用于根据预设目标状态确定量子黑盒。
进一步地,确定单元204,还用于通过以下公式,根据预设目标状态确定量子黑盒,
O=I-2|τ><τ|。
式中,O为量子黑盒,I为单位矩阵,|τ>为预设目标状态。
本实施例未尽之细节,请参阅前述图1至图5所示的实施例的描述,此处不再赘述。
在本发明实施例中,通过利用多变迭代算子,使得后续迭代运算后使用的迭代算子随着迭代次数的增加而变换,多变迭代算子的运用加快了目标状态概率幅的增加,加快了非目标状态概率幅的降低。并且在几何表示中,旋转轴同样随着迭代次数的增加而变换,旋转角度为3
kθ,而不是利用固定的旋转轴,从而突破了原有量子搜索算法计算速率的限制,提高了搜索被标记的量子状态的计算速率。
在上述实施例中,对各个实施例的描述都各有侧重,某个实施例中没有详 述的部分,可以参见其他实施例的相关描述。
以上为本发明所提供的量子状态搜索方法及装置的描述,对于本领域的一般技术人员,依据本发明实施例的思想,在具体实施方式及应用范围上均有改变之处,综上,本说明书内容不应理解为对本发明的限制。
Claims (10)
- 一种量子状态搜索方法,其特征在于,所述方法包括:根据哈达玛变换效应将量子初始态进行变换,得到量子输入态;根据量子黑盒和所述量子输入态确定迭代算子,基于所述迭代算子和所述量子输入态上进行迭代运算,得到量子中间态,并将所述量子中间态确定为新量子输入态;执行所述根据量子黑盒和所述量子输入态确定迭代算子,基于所述迭代算子和所述量子输入态上进行迭代运算,得到量子中间态,并将所述量子中间态确定为新量子输入态的步骤,直至达到预设的终止条件,则得到的量子中间态确定为被标记的量子状态。
- 如权利要求1所述的量子状态搜索方法,其特征在于,所述量子黑盒为酉矩阵形式,则通过以下公式,执行所述根据量子黑盒和所述量子输入态确定迭代算子,基于所述迭代算子和所述量子输入态上进行迭代运算,得到量子中间态,并将所述量子中间态确定为新量子输入态的步骤,直至达到预设的终止条件,则得到的量子中间态确定为被标记的量子状态:|ψ k>=G k-1G k-2…G 1G 0|ψ 0>;G i=(2|ψ i><ψ i|-I)O;式中,|ψ k>为被标记的量子状态,|ψ i>为第i+1次迭代运算中使用的量子输入态,G i为i+1次迭代运算中使用的迭代算子,i∈[0,k-1],k为迭代终止次数,i∈[0,k-1],O为所述量子黑盒。
- 如权利要求3所述的量子状态搜索方法,其特征在于,所述量子输入态为等幅度值的叠加态,则所述根据哈达玛变换效应将量子初始态进行变换,得到量子输入态之后,包括:根据预设目标状态确定量子黑盒。
- 如权利要求4所述的量子状态搜索方法,其特征在于,通过以下公式,根据预设目标状态确定量子黑盒:O=I-2|τ><τ|;式中,O为量子黑盒,I为单位矩阵,|τ>为预设目标状态。
- 一种量子状态搜索装置,其特征在于,所述装置包括:变换单元,用于根据哈达玛变换效应将量子初始态进行变换,得到量子输入态;迭代单元,用于根据量子黑盒和所述量子输入态确定迭代算子,基于所述迭代算子和所述量子输入态上进行迭代运算,得到量子中间态,并将所述量子中间态确定为新量子输入态;输出单元,用于执行所述根据量子黑盒和所述量子输入态确定迭代算子,基于所述迭代算子和所述量子输入态上进行迭代运算,得到量子中间态,并将所述量子中间态确定为新量子输入态的步骤,直至达到预设的终止条件,则得到的量子中间态确定为被标记的量子状态。
- 如权利要求6所述的量子状态搜索装置,其特征在于,所述量子黑盒为酉矩阵形式,则,所述输出单元,还用于通过以下公式,执行所述根据量子黑盒和所述量子输入态确定迭代算子,基于所述迭代算子和所述量子输入态上进行迭代运算,得到量子中间态,并将所述量子中间态确定为新量子输入态的步骤,直至达到预设的终止条件,则得到的量子中间态确定为被标记的量子状态:|ψ k>=G k-1G k-2…G 1G 0|ψ 0>;G i=(2|ψ i><ψ i|-I)O;式中,|ψ k>为被标记的量子状态,|ψ i>为第i+1次迭代运算中使用的量子输入态,G i为i+1次迭代运算中使用的迭代算子,i∈[0,k-1],k为迭代终止次数,i∈[0,k-1],O为所述量子黑盒。
- 如权利要求8所述的量子状态搜索装置,其特征在于,所述量子输入态为等幅度值的叠加态,则所述装置还包括:确定单元;所述确定单元,用于根据预设目标状态确定量子黑盒。
- 如权利要求9所述的量子状态搜索装置,其特征在于,所述确定单元,还用于通过以下公式,根据预设目标状态确定量子黑盒:O=I-2|τ><τ|;式中,O为量子黑盒,I为单位矩阵,|τ>为预设目标状态。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2018/110389 WO2020077521A1 (zh) | 2018-10-16 | 2018-10-16 | 量子状态搜索方法及装置 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2018/110389 WO2020077521A1 (zh) | 2018-10-16 | 2018-10-16 | 量子状态搜索方法及装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020077521A1 true WO2020077521A1 (zh) | 2020-04-23 |
Family
ID=70282902
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2018/110389 Ceased WO2020077521A1 (zh) | 2018-10-16 | 2018-10-16 | 量子状态搜索方法及装置 |
Country Status (1)
| Country | Link |
|---|---|
| WO (1) | WO2020077521A1 (zh) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7028275B1 (en) * | 2001-10-15 | 2006-04-11 | The Texas A&M University System | Quantum circuit design for grover's algorithm |
| CN102495886A (zh) * | 2011-11-28 | 2012-06-13 | 河南理工大学 | 基于量子算法的指纹数据库搜索方法 |
| CN107025206A (zh) * | 2017-04-13 | 2017-08-08 | 广西师范大学 | 一种量子傅立叶变换实现量子线路设计的方法 |
-
2018
- 2018-10-16 WO PCT/CN2018/110389 patent/WO2020077521A1/zh not_active Ceased
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7028275B1 (en) * | 2001-10-15 | 2006-04-11 | The Texas A&M University System | Quantum circuit design for grover's algorithm |
| CN102495886A (zh) * | 2011-11-28 | 2012-06-13 | 河南理工大学 | 基于量子算法的指纹数据库搜索方法 |
| CN107025206A (zh) * | 2017-04-13 | 2017-08-08 | 广西师范大学 | 一种量子傅立叶变换实现量子线路设计的方法 |
Non-Patent Citations (1)
| Title |
|---|
| LI, SHIYONG ET AL., QUANTUM COMPUTATION AND QUANTUM OPTIMIZATION ALGORITHMS, 31 May 2009 (2009-05-31), pages 46 - 50 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107025206B (zh) | 一种量子傅立叶变换实现量子线路设计的方法 | |
| WO2022077646A1 (zh) | 一种用于图像处理的学生模型的训练方法及装置 | |
| CN110738321A (zh) | 一种量子信号处理方法及装置 | |
| CN111680504B (zh) | 法律信息抽取模型及方法及系统及装置及辅助系统 | |
| WO2021232752A1 (zh) | 哈希编码方法、装置、设备及可读存储介质 | |
| CN116113941A8 (zh) | 一种神经网络加速器、加速方法以及装置 | |
| WO2016062044A1 (zh) | 一种模型参数训练方法、装置及系统 | |
| WO2021218480A1 (zh) | 基于模拟量子算法的数据搜索方法、装置及设备 | |
| CN111815738A (zh) | 一种构建地图的方法和装置 | |
| CN109214517A (zh) | 量子状态搜索方法及装置 | |
| CN106599311A (zh) | 一种基于云计算的互联网教育平台资源库的图像检索方法 | |
| CN112163145B (zh) | 基于编辑距离与余弦夹角的网站检索方法、装置及设备 | |
| CN112686342B (zh) | Svm模型的训练方法、装置、设备和计算机可读存储介质 | |
| CN115641440A (zh) | 一种点云的语义分割方法、装置及存储介质 | |
| Li et al. | Quantum support vector machine based on gradient descent | |
| US12468973B2 (en) | Physical transformations for multi-dimensional data quantum representations | |
| Han et al. | IDPD: improved deformable-DETR for crowd pedestrian detection | |
| US20130246462A1 (en) | Autonomic caching for in memory data grid query processing | |
| Yin et al. | Nonlinear low‐rank representation on Stiefel manifolds | |
| WO2023024210A1 (zh) | 一种基于傅里叶域主成分分析的数据降维方法 | |
| CN111339724A (zh) | 用于生成数据处理模型和版图的方法、设备和存储介质 | |
| CN107330929A (zh) | 一种基于几何重心和质心距离比不变性的多尺度点云配准方法 | |
| Yang et al. | Homotopy relaxation training algorithms for infinite-width two-layer relu neural networks | |
| Liu | An exact quantum search algorithm with arbitrary database | |
| CN110135340A (zh) | 基于点云的3d手部姿态估计方法 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18937258 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 18937258 Country of ref document: EP Kind code of ref document: A1 |








